Full text of "Anthropological essays presented to Edward Burnett Tylor in honour of his 75th birthday, October 2, 1907"

See other formats

This is a digital copy of a book that was preserved for generations on library shelves before it was carefully scanned by Google as part of a project
to make the world's books discoverable online.

It has survived long enough for the copyright to expire and the book to enter the public domain. A public domain book is one that was never subject
to copyright or whose legal copyright term has expired. Whether a book is in the public domain may vary country to country. Public domain books
are our gateways to the past, representing a wealth of history, culture and knowledge that's often difficult to discover.

Marks, notations and other marginalia present in the original volume will appear in this file - a reminder of this book's long journey from the
publisher to a library and finally to you.

Usage guidelines

Google is proud to partner with libraries to digitize public domain materials and make them widely accessible. Public domain books belong to the
public and we are merely their custodians. Nevertheless, this work is expensive, so in order to keep providing this resource, we have taken steps to
prevent abuse by commercial parties, including placing technical restrictions on automated querying.

We also ask that you:

+ Make non-commercial use of the files We designed Google Book Search for use by individuals, and we request that you use these files for
personal, non-commercial purposes.

+ Refrain from automated querying Do not send automated queries of any sort to Google's system: If you are conducting research on machine
translation, optical character recognition or other areas where access to a large amount of text is helpful, please contact us. We encourage the
use of public domain materials for these purposes and may be able to help.

+ Maintain attribution The Google "watermark" you see on each file is essential for informing people about this project and helping them find
additional materials through Google Book Search. Please do not remove it.

+ Keep it legal Whatever your use, remember that you are responsible for ensuring that what you are doing is legal. Do not assume that just
because we believe a book is in the public domain for users in the United States, that the work is also in the public domain for users in other
countries. Whether a book is still in copyright varies from country to country, and we can't offer guidance on whether any specific use of
any specific book is allowed. Please do not assume that a book's appearance in Google Book Search means it can be used in any manner
anywhere in the world. Copyright infringement liability can be quite severe.

About Google Book Search

Google's mission is to organize the world's information and to make it universally accessible and useful. Google Book Search helps readers
discover the world's books while helping authors and publishers reach new audiences. You can search through the full text of this book on the web

at |http : //books . google . com/

PRESENTED TO

EDWARD BURNETT TYLOR

OCTOBER 2 1907

HENRY FROWDE, M.A.

PUBLI8BBB TO THE UNITSBSITY OF OXFORD

LONDON, EDINBtJBOH

NEW YORK AND TOBONTO

ANTHROPOLOGICAL ESSAYS

PRESENTED TO

EDWARD BURNETT TYLOR

IN HONOUR OF HIS 75th BIRTHDAY OCT. 2 1907

H. BALFOUR A. E. CRAWLEY D. J. CUNNINGHAM L. R. FARNELL

J. G. FRAZER A. C. HADDON E. S. HARTLAND A. LANG

R. R. MARETT C. S. MYERS J. L. MYRES C. H. READ

SiE J. RHtS W. RIDGEWAY W. H. R. RIVERS

C. G. SELIGMANN and T. A. JOYCE

N. W. THOMAS A. THOMSON

E. WESTERMARCK

WITH A BIBLIOGRAPHY BY --..
BARBARA W. FREIRE-MARREGa'i

OXFORD
AT THE CLARENDON Pl^ESS

1907

OXFORD

PRINTED AT THE CLARENDON PRESS

BY HORACE HART, M.A.

PRINTER TO THE UNIVERSITY

•••••

•

• • •

• •

•

'•v,V

• •••,

• •

• • •

• ••••

• • •

• • J

• ••

*•*•*,','

.....

•*.'.'.

• ,•

• ••••

•••:.

V.**'.

• •

t « •

;;V/

• •

.••:*

• •.

•• .

• •

• • ••

•••

••••.

• • •

• •

• • •

1118S6

..> •.-

PREFACE

Early in the present 5^ear it was resolved to com-
memorate the seventy-fifth birthday of Dr. E. B. Tylor
by presenting to him a vohiine of anthropological
essays. If the volume is not entirely representative of
English anthropology at the present day, the cause is
to be sought in the short period available tor the
prei)aration of the essays and in the desire of all to
give nothing which might seem to fall short of their
best work. But in this ottering are associated both
those who contribute to the volume and others who,
from lack of opportunity, were unable to lay a gift
before the greatest of English anthropologists.

W. H. R RIVERS .

R R MARETT | Editorial

I Committee.
NORTHCOTE W. THOMAS )

The responsibility for the collection of the essays

was confided to the committee whose signatures

appear above, but the actual work of editing the volume

and seeing it through the press has lain almost entirely

in the able hands of Mr. Thomas.

W. H. R R

R. R M.

CONTENTS

EDWARD BURNETT TYLOR. page

By Andbbw Lano, M>A.., LL.D 1

HENRY BALFOUR, M.A.

The Fiee-Piston. With Map and Plates I-IV . 17

A. E. CRAWLEY, MJ^., FJI.A.L

Exogamy and the Mating of Cousins .... 51

D. J. CUNNINGHAM, M.D., D.Sc, LL.D., D.C.L., F.R.S .

The Australian Fobehead. With Plates V-VII 65

L. R. FARNELL, D.LITT.

The Place of the ' Sond£B-(3ott£b ' in Gbeek Poly-
theism 81

J. G. FRAZER.

Folk-lobe in -the Old Testament . . 101 ^

ALFRED C. HADDON.

The Religion of the Tobbes Stbaits Islandebs ^ . . 175

E. SIDNEY HARTLAND, F.S.A.

CONOEBNINO THE RiTE AT THE TeMPLE OF MyUTTA ^ . 189

ANDREW LANG, MJL, LL.D.

Austbalian Pboblems 203

R. R. MARETT.

Is Taboo a Negative Magic? 219

CHARLES S; MYERS, M.A;, M.D.

The Ethnological Study of Music 235

viii CONTENTS

J. L. MYRES. PAOE

The Sioynnae of Hebodotds: An Ethkolooical Pboblem

OF THE Eablt Iron Aoe 255

C. H. READ, F.S.A.

A Museum of Anthbopolooy. With Plan .... 277

Sm JOHN RHYS.

The Nine Witches op Qlouoesteb 285

— WILLIAM RIDGEWAY. M.A., F.B.A., Hon. D.LiTr.

Who webe the Dobians? 295

w. h. r. rivers.

Os the OBtaiN OP THE CliASSIPICATOBY SYSTEM OP RELA-
TIONSHIPS ......... 309

C. G. SELIGMANN, M.D., and T. A. JOYCE, M.A.

On Pbehistobic Objects in Bbitish New Guinea. With

Plates VIII-XIII and Pigs, 1-5 in text . .326

NORTHCOTE W. THOMAS, M.A.

The Obiqin of Exoqaicy 343

ARTHUR THOMSON, M.A.

The Secbet of the Veboe Watch : A Study in Symbolism

AND Desiqn. With Plates XIV, XV . .355

EDWARD WESTERMARCK, Ph.D.

L'Ar, ob the Tbansfebence of Conditional Cubses in

MOBOCCO 361

A BIBLIOGRAPHY OF EDWARD BURNETT TYLOR.

By Babbaba W. Fbeibe-Mabbeco 375

INDEX 410

EDWARD BURNETT TYLOR

By ANDREW LANG, M.A, LL,D.

It was my fortune to make Mr. Tylors acquaintance at
Oxford, about the year 1872, before I had heard of any of his
books, in which his masterpiece, Primitive Culture, was already
numbered. The distinguished and witty lady who introduced us
to each other had, and, alas! has, rather more than the ordinary
aversion from things primitive, and from the study of benighted
heathens. She informed me that Mr. Tylor had written *a large
book, all about savages', in whom Mr. John Fergus M'Lennan
had already interested me by liis essays on totemism, Thus my
acquaintance with Mr. Tylor and his great book began thirty-five
years ago, when he, beside Sir John Lubbock, already towered above
all British anthropologists, like Saul above his people. Since these
early days I have often had the pleasuie of being with Mr. Tylor in
social fashion, and have again and again perused his books. But it
is my misfortune to know little of his Museum work, though even
brief and cursory visits to the Pitt-Rivei*s Museum demonstrate that
it is on a level of excellence with his written expositions ; and I have
never been present at any of his lectures to his Oxford pupils. His
later yeai^ have been spent in academic toils ; he has sent his pupils
into many strange lands ; they have been the field naturaUsts of
human nature, no less than anthropologists of the study. If England
possesses an unofficial school of antlnopologists, despite the pubhc
mdifference to man not fully * up to date ', she owes it to the examples
of Mr. Tylor and Lord Avebury. But I am only an amateur, and have
especially to deplore my slender acquaintance with the work of Mr.
Tylor s eminent German predecessors and contemporaries. A pupil
less competent than I to estimate Mr, Tylor s work in its relations
to his study, as pursued on the Continent and in America, could
scarcely be found. Indeed, I speak rather as one of the outer circle
— of the Court of the Gentiles — than as a professed antliropologist.

It is to be noted that, in 1860-1870, a fresh scientific interest in
matters anthropological was * in the air \ Probably it took its rise>

TTlX»lt B

2 EDWARD BURNETT TYLOR

not so mucli in Darwin*s famoizs theory of evolution, as in the long-
ignored or ridiculed discoveries of the relics of Palaeolithic man by
M, Boucher de Perthes. Mr. Henry Christy, a friend of Mr. Tylor,
and Sir John Evans, helped greatly to establish the authenticity of
the discoveries of M. Boucher de Perthes, and while they were mainly
exercised with the development of man's weapons, implements, and
arts, Mr. Tylor, with Lord Avebury, studied his mental development
^as revealed in liis customs, institutions, and beliefe. Mr. McLennan
and Sir Henry Maine were contemporaneously laying the foundations
of the study of earUer and later jurisprudence, and of this generation
of heroes we are but the epigonoi ; fortunate in this, that we still
have among us so many distinguished survivors of the great age.

The track or trail left by our ancestors of the stone age has for
thousands of years attracted curious minds. Hesiod had his theory
of progress and of successive races, beginning with gods, followed by
heroes, and passing through the age of bronze, * when as yet black
iron was not,* Moschion touches on cave-dwellers, whom he regards
as cannibals ; and Lucretius traces rehgion to the belief in spirits, or
* animism ', bred of reflection on the phenomena of breath, dreams,
and shadows. Tlie Greek geographers, and Herodotus and Aristotle,
were curious about the institutions of savage and barbaric races ;
wliile, in the eighteenth century, Goguet, Fontenelle, Boulanger,
des Brosses, Professor Millar, of Glasgow, and othere, explained the
rise of mjrthology, and the origin of rank, on the lines of modern
anthropological science. The idea of evolution, for aU that we know,
is as early a conception of thmking men as the idea of creation ; both
exist among the most primitive savage races ; and, in short, all that
the speculators of the last and the present age can do is to bring
wider study, and more precise methods, into the investigation of
human development.

In the middle of tlie nineteenth century the advance of philo-
logical science, with the theory that mythology is the result of * decay
of language * ; and the other theory that degeneration has more to
answer for than we can admit, caused a temporary diversion from
the ideas of Lucretius and Fontenelle. Fortunately the.se notions did
not distract Mr. Tylor from the path which he was born to follow.
His interest in his subject may have been aroused by the early tour
t« Mexico which bore fruit in his fii-st volume Anahtuic^ or Mexico
and the ^lexicam (1861), followed by his volume Researches into the
Early History qf Mankind (1865, 1870, 1878).

EDWARD BURNETT TYLOR

It is interesting and instructive to look back at this work of
forty years ago. It is a series of essays towards a history of civiliza-
tion, a history necessarily based rather on Reulien^ savage weapons,
implements, arts and crafts, and on myths, customs, and beUefs, than
on written materials. Forty years ago, Mr. Tylor was conscious
of working on a new topic ; but he has made it famihar to some
members of two generations of Englishmen ; in Germany much

* culture history ' had already been written.

Mr, Tylor's main interest has been in belief and institution,
but at a later date he made a notable contribution to the study of
Realien by his article {J. A, L^ Xs 74) on the plough and wheel-
carriage. Ten years after this (J. A, L, xix. 54) he traced the 'face-
brasses * on harness to their source in the pJmlerae of the Romans.

* From beginning to end of his book, the author*s mind was
occupied by the question, among the countless coincidences of
customs, beliefs, arts and crafts, games, riddles, proverbs, institu-
tions, how much has been diffused and borrowed, how much is of
independent invention ? Did Aztecs and Polynesians borrow from
Asiatic sources? When a tribe (probably now extinct) on the
Glenelg river in Victoria, polished stone axes of jade, so rare a mineral
in AustraUa, had they learned the art from Malays? Or is the
explanation that, finding unusualiy good material, they worked it
with unusual care ? Here is a problem in the solution of which we
have made little progress in forty years. My own bias is to seek a
solution in original, independent, and coincidental invention ; while,
even now, popular writei-s lean to theories of unity of race ; and even
totemism has been regarded as diffused from some single unknown
centre. Mr. Tylor, in The Early History of Mankind^ tended more to
a theory of the borrowing of myths and Mdrchen than I have ever
been disposed to do ; in short the natural bias of the speculator
usually affects his opinion in this diflScult case, except when there is
distinct and definite evidence for a single original centre of invention.
Tliat spiritual or animistic beUefs arise, independently, wherever
men dream by night, and see phantasms by day, Mr. Tylor, in 1865,
already maintained, developing the idea into a theory of the origin
of reUgion, in his later work, Primitive Ctdture. He was certain that
phantasms beheld mth waking eyes are * subjective processes of the
mind', and did not trouble himself about * coincidental ' and

* veridical ' apparitions, till he wrote Primitive Cidture, The materials
which have come in since 1865 afford many additions to his excellent

B 2

SDWAKD BCBSETT TYI/OfSL

IRrmtfn Cretan and Lefaotiiie diaeovcns luve eonfeiifaiitod mndi to
ii^U>pie(A'Tk!taieWntiDgMkiWagdWnidM^\ In iUB field the
htAt&eeajeanhssweheeaftefoJiaii^ Butnodiiiig

has been dkeorend m to the mflnftire of ^SameB^widdiinB not
irtatedy or Ibfeahadowed in Mc: Tflor^s stodf of die atdgeet, and his
diapter about ^Gfowth and Dedine in Coitnre' Snaij tneed the
fines on wfaidi sei^iee is still content to build, ffis ch^iters on
'The Stone Age, Past and Present 'are stiU the best En^iah intro-
dneti<m to the subject. The pages on 'Kre and Cooking' oq^ not
to be nej^tected by the sweet enthnsiaHts who peiaevere in avening
that Pc^rnesians cannot fi^it fire, and do dieir cookii^ without it

Atopic in wfaidi forty years have sem mndi fitoh knowledge
garnered is what Mr. Tylor called 'The Comparative Jan^arndenee
of the lower races '. His evidoiee was collected before the paUiea-
tion of J. F. M^Lennan's epocb^nakii^ ^PrimitiYe Marriage', which
Mr. Tylor safaited as ^the first qratematic and acientific attaoopt to
eficit general ininciples from the diaotic mass of detaib of savage
law. . . / Chaotic they still rmiained, fi»r we find that the word
'tribe' was then (as it sometimes is still) naed as a synonym for
'totem-kin 'y and also for 'the matrimonial classes', or 'sobdasses'
of die Kamilaroi I^mo, Kubbi, Knmbo, and Bnta af^pear as ' tribes ',
reported on mider that tide by I>r. Lang : and, with Sir George Grey ,
the totem name is 'the fomily name'. The IroqiMHS hare 'ei^it
tribes', these 'tribes' being really 'toteubkins'; in short, forty
years ago, information was scanty, and terminology was eren more
indistinct than, unhappily, it is at present There was better
information about ' Avoidances ' between kinsfolk wh^her by blood
or 'in law', but the original purpose of these ayoidances is still
matter of controversy. Why, for example, may 'the fother not
f^>eak to his son after his fifteenth year'? It does not appear
probable that all avoidances were instituted for the same reason,
and Mr Tylor found no single reason that ¥^uld account for all
avoidances.

The researches and speculations of almost half a century, into
some parts of the jurisprudence of the lower races, have foiled to
produce any agreement of opioion, at least in the case of laws
regulating marriage, but the remarkable statement has been emitted
that we ought especially to distrust any hypothesis which, in complex
matter, professes to colligate all the foots. The hypotheses which

SDWARD BUENETT TYLOR

fail to do so are the more respectable. Of these, happily, there is
great plenty.

For the next four yeara after 1865, Mr. Tylor was laying the
foundation stones of his system in papers on -The Eeligion of
Savages', *The Condition of Preliistoric Races,' *The Survival of
Savage Thought in Modem Civilization/ and other essays. At the
close of this period the researches of Mr, Lewis Morgan into
systems of kindred, with the very original pioneer essays of
Mr, J. F, M*Lennan on Totemism in relation to Primitive Marriage
and Exogamy, opened a field as thorny as expansive, a field into
which Mr. Tylor, as far as his published works are concerned, has
made few incursions.

The most notable of these is his epoch-making article {J, A, J.,
xviii. 245) on a Method of Investigating the Development of Institu-
tions appUed to the Laws of Marriage and Descent. In tliis Mr< Tylor
aimed at showing that the development of institutions might be
investigated on a basis of tabulation and classification. He had
scheduled out into tables the rules of marriage and descent all the
world over, so as to ascertain what may be called tlie * adhesions ' of
each custom, showing what people have the same custom and what
other customs accompany it or he apart from it.

On this basis Mr. Tylor discussed the rules of residence after
marriage, its connexion with avoidance, with teknonymy (naming
of parents after children), with the levirate, with the couvade, and
with marriage by capture, and showed that the residence of the
husband with the wife's people was, so ftir as his schedulas gave
information, indisputably anterior to the residence of the wife with
the husband, though he was careful to point out that this was not
necessarily the most primitive state of things,

Tlien taking up exogamy and the classificatory system of
relationship, Mr, Tylor displayed them as two sides of the same
system, and argued that the purpose of exogamy was to enable
a growing tribe to keep itself compact by constant unions between
its spreading clans. Finally he stated that there were still a hundred
or more peoples in the world for which he had no information, and
expressed the hope that each civihzed country would take in hand
the barbaric regions within its purview.

In 1871 he produced his chief work, Primitive Culture^ and at
once appeared as the foremost of British anthropologists. The
extent of his reading, his critical acumen, his accuracy, his power

EDWAED BURNETT TYLOR

of expoeitioDf Im open mindf and his scientific caution make this
book no paaaing eesay, but a poBsession for ever. He laid the firm
fbctodation of a structure to which, with aocniing information, others
may make additions ; he himself has made and is making additions ;
but hiH science passed, thanks to him, out of the pioneering stage, at
a single step. He stood on a level with Bastian ; their names are,
in the pre-historic history of man, what the name of Darwin is in
regard to the evolution of animal life. There are, indeed, as there
muiit be, modifications to be suggested, and verdicts to be revised ;
but in the future, as in the present, it is from Mr. Tylor's work that
tha beginnings must be made; and he who would varjr from
Mr. Tylor's ideas must do so in fear and trembling (as the present
writer knows by experience).

Mr. Tylor, in the preface to his second edition (1873), observes
that ' writers of most various philosophical and theological schools
now admit that ethnological facts are real, vital, and have to be
accounted for '. He had emancipated us from exclusive and rather
fanciful attention to * the Aryan race '. He had proved that man,
in Byron's words, is * always and everywhere the same unhappy
fellow', whatever the colour of his hair or skin, and the shape of
his skull Ifomo, in the earliest stage at which we make his
acquaintance, is already the philosopher, artist, and man. He

* linds something craggy to break his mind upon ', and we have
scarcely a tlieory concerning the deeper problems of life which
savage man has not already invented in his mythical Platonic way.
E*ich of his niytlis, for example, explaining the origin of Totemism
has its counterpart in a modern theory : liis ghosts are our phantasms ;
and his religion justifies a famous saying of TertuUian. We cannot
escape from him in any field of activity; we repeat his theories
without knowing ; or knowingly, as when Mr. F. W, H. Myers boldly
proclaimed his own rovei'sion to 'palaeolithic psychology *, Without
the ideas of the savage (as Keats averred) we should have no poetry
worthy of tho name, and these fruitful rudiments, not to be styled

* superstitions ', Mr, Tylor named * survivals ' ; a term which implies
no R^proacli,

But it is fair to civilized man to say, that if his savage ancestors
had bequeathed to him no supei'stitions, he would have invented
tliom anew for himself, Sucli is human nature ; witness the cases
of Zola and Dr. Johnson,

Not tho least of Mr. Tylor s gifts, as the founder of his science,

EDWAED BURNETT TYLOR

is the happy simplicity and unobtrusive humour of his style. Not
stu£Fe<l with strange technical words, his language, as in his admirable
chapter on * Survival in Culture ' (iii) is so attractive, so pellucid, that
any intelligent child could read it with pleasure, and become ay
folk-lorist unawares. The doctrine of survivals, though incontestable
in general^ has its difficulties. We meet phenomena in savage
culture which one set of students recognizes as ' survivals ' ; while, in
the same facts, other inquirers see novelties, freaks, or ' sports \ An
example is familiar ; several of the customs and beliefs of the tribes
of Central Australia are, on one side, explained as survivals of
primitive^ on the other, as recent modifications of decoder^ totemism.

From survivals in games, proverbs, riddles, and the minor
superstitions, such as those of sneezing, Mr. Tylor glides into
Magic, as based on association of ideas ; into omens, automatisms,
witchcraft, spirituaUsm, and the doctrine of spirits, ' Animism,* with
its influences on religion and mythology. Even races wliich believe
in magic, he says, unconsciously judge it when they regard their
more backward neighbours as more potent magicians than them-
selves. Protestants in Germany, says Wuttke, get Catholic priests
to lay ghosts for them. Why not, if the ghost be a Catholic ghost ?
The Kev. Mr. Thomson of Ednara, father of the author of The Castle
of Indolence^ was slain by a ghost, obviously not Presbyterian, whom
he, a Presbyterian, imprudently attempted to lay. The house
haunted by this ghost had to be pulled down, so say the annals of
the parish.

Nothing stands still, and, since the date of Mr. Tyler's book,
psychologists have studied some savage modes of divination, for
example by the divining-rod, and ciystal-gazing, as instances of

* automatism *, and of the action of the sub-conscious self. The
divining-rod, not known to the Austrahan black water-seeker,
survives among oiu^elves, because the automatic faculty survives
in man, even when he has science enough to explain the phenomena
not by the agency of spirits, but * electricity ^ Melanesians and
other savages have observed facts of automatism, motor or sensory,

^and explained them, of coui'se, by the action of spirits. It is the
inimistic explanation, as held by modern spirituahsts, not the facts
of automatism, that is the sui-vival. Mr. Tylor concluded that

* there is practically no truth or value whatever' in *the wiiole
monstrous fan^ago'. But there now appears to be (indeed Mr.
Tylor himself foresaw the fact) a good deal of value for the

EDWABD BURNETT TYLOE

peychologist, and some light lb thrown on the more obscure faculties
of the race*

Mr, Tylor himself recognized that * occult* arts may produce
practical effects by suggestion, a fact noted, utilized, and erroneously
explained by savages and rustics, . and even by Richard Baxter,
Woodrow, the Mathers, and other learned divines. The frenzy
of spiritualism, as Mr. Tylor showed^ in the mid nineteenth centmy,
was a revival, or recrudescence in culture. It resembled the jurore
against witches, which, north of Tweed, came in imder John Knox,
and flourished through the Puritan period; though, in Scotland,
the elder faith had been too wise, or too indolent, to persecute
'witches. With his habitual caution and open-mindedness, Mr,
Tylor remarked that a careful and scientific observation of some
of the new or revived marvels * would seem apt to throw light
on some most interesting psychological questions', beyond the
scope of his inquiry. An instance in point is the 'Poltergeist*,
though it cannot be said that observation lias done much to explain
Aim, except as associated with the presence of a more or less
* liysterical * person* Mr. Tylor s affair was to discover great numbers
of ethnological parallels to the speciosa miracula of spiritualism,
and to leave the matter there for the present; while the savage
animistic explanation led up to the whole vast subject of Animism,

A most interesting part of Mr. Tylors work is his analysis
(chapter VI) of words denoting human relationships, and their
connexion with ' baby language ', If we follow the linguistic indica-
tions, fatherhood seems of as early recognition as motherhood. We
are not told in what tribal language of Australia viammun means
rfather', and, in such lists as Messrs, Spencer and Gillen and
Mr, Howitt supply, words for 'father' derived from *baby lan-
guage' do not seem to prevail. The papa of the Carib and the
Caroline Islander and the modem Briton, when it occurs in seveml
North Australian languages, means the seniors, but not the father.
The *baby language' terms give 'striking proof of the power of
consensus of society, in establishing words in settled use mitlmti
their camjing traces of inherent expressiveness*.

It does not foUow, I may add, that such inexpressive terms
of relationship imply a past when men did not recognize con-
sanguinity. Mr. Tylor, in 1871, did not judge the time ripe for
a discussion of classificatoiy terais of relationship, of early marriage,
of exogamy, and totemism. On these topics we may expect, and

EDWARD BURNETT TYLOR

even witli impatience, his mature views in the great work with
which he has long been occupied.

in 1871 he spoke, as concerns totemism, of 'the direct worship
of the animal for itself, indirect worship of it as a fetish acted
tlirough by a deity, and veneration for it as a totem or repre-
sentative of a tribe-ancestor'. He also connected the totem with
animism, the worship of *a divine ancestral soul' (pp. 237, 238),
A leaning towards the same theory may be observed in his ' Remarks
on Totemism' (1898, J. A. X, xxix. 138). Tlie difficulties in the
way of this hypothesis have often been pointed out, and we expect
their solution.

From the first, in 1871, Mr, Tylor distinguished sharply between
the totem of the kin, hereditary in the female or male line, and
*the mere pati*on animal of the individual'. This essential dis-
tinction he has continued to maintain. On the whole topic Mr.
Tylor has ever shown great and laudable caution ; may others be
forgiven who have hazaided hjrpotheses much at the mercy of new
invading facts tliat undermine our cloud-capped towers of conjecture !
Mr. Tylor rejected the explanation of totemism, which derived it
from the adoption, by a man*s descendants, of his individual name,
such as Bear, Deer, or Eagle. It would, in fact, be necessary
to substitute here for ' the individual man ', ' the individual woman,'
among tribes which inherit the totem on the female side.

Arriving at the old problem of the origins of mythologies,
Mr. Tylor fell back on the ancient wisdom of Eusebius of Caesarea,
the half forgotten sense of Fontenelle ; took * savage mythology
as a basis ', and convincingly proved that mytliology is the natural
product of the mental condition of savages.

Even in Greece, in tales usually left untold or carefully sub-
ordinated by Homer, myth retains its savage birth-marks. With
this key the old lock is opened, and we understand that the mythical
vagaries of gods, and beasts, and men, closely resembling even
in minute details the stories of savages, are survivals, repulsive
flies caught in the amber of ritual and reUgion. This theory
Mr. Tylor worked out with wonderful tact, never throwing a stone
into the adjacent garden of Mr. Max Mtiller, whose solar theory
and philological method were then dominant in this country.
Mr. Tylor's idea was not new ; perhaps there are no new ideas j
his merit lay in his patient, sagacious, well * documented', and,
at last, convincing method of exposition. Nothing was left but

EDWAED BURNETT TYLOR

to apply the system in detail to every realm of mythological
creation; and though, in certain learned circles at home and
abroad, the method was for long ignored, or resisted, in the end
it has triumphed; accompanied, as it has been, by the system
of Mannhardt, who paid more attention to European folk-lore, with
its survivals of early ritual, than did Mr, Tylor,

The scent may, of course, be overrun by too eager pursuers.
It seems rather hasty to maintain that the tuneful Orpheus ■ was
a fox, a * sex-tot^m' of the women of Thrace ; and I sympatliize
with the cautious author who remarked that * Blindnmn's Buff' is not
necessarily a survival of human sacrifice'. It is possible, or rather
it is easy to consider too curiously; there was a tendency to see
totems everywhere, as in the name and crest of Clan Chattan, or
in the final syllable of Glencoe, which a judicious linguist will
not translate 'the Glen of the Dogs*. There was a time when
I was apt to see churinga nanja everywhere ; and my sole excuse
is that the European neoHthic and palaeolithic things were exactly
like churinga nanja. We generally have some special pet idea
%vhich we overdo ; not taught by the reserve of Mr, Tylor, whom
a kind nature has exempted from the obsession of the idolon specm^
and whose method does not lend itself to parody.

If this were the place for criticism, and if I were anxious to
' lift up my hands against my father Parmenides ', I would confess
a certain difficulty. Is * the belief in the animation of nature, rising
at its highest pitch to personification * — and, in itself, a main source
of mythology — identical with *the doctrine of Animism' (i, 285)?
Could a savage, or a child, not conceive (in the spirit of analogy)
that the wind or the sun is, like himself, a living person, before the
child had heard of a ghost, or the savage had developed for himself
the belief in bodiless souls ? It appears that the notion of universal
vitality is, really, not the same as ' the doctrine of Animism, whicli
develops and reacts upon mythical personification' (i. 287). In-
deed, as Mr. Tylor observes, many forms of thought *work in
mythology with such manifold coincidence as to make it hard
indeed to unravel their separate action \

Animism, indeed, is first treated apart from Mythology; and
the Lucretian theory of the origin of spiritual beliefs is worked out
through a long sequence of examples. But Lucretius had to adopt
a theory to account for the casual hallucinatory phantasms beheld
by sane, healthy, and waking men, which was easily demolished by

EDWARD BURNETT TYLOR

Plutarch ; while Mr. Tylor is too wary * to discuss on their merits the
accounts of what is called second sight \ Savages as well as civilized
persons have the second sight, and that is enough for liis purposes.
He gives modem instances — which are as common as blackberries —
but it suffices him that these experiences ofifer one basis of the doctrine
of Animism, and it is not his business to ask whether the basis is
not a pretty solid foundation stone for a towering palace of miraga

In Animism the savage philosopher had a ready key to most
problems that puzzled him. - Spirits were * at the bottom of them * ;
they were, to him, what electricity is to the modern popular mind.
It is even more curious to notice how much savages differ in their
animistic philosophies than to observe the points in which they
agree. It seems to myself that, except in East Africa, Fiji, and,
doubtless other regions, the savage, or barbarian, is not much of
a ghost-seer. The Masai are said not to believe in ghosts, though,
inconsistently, they say that cattle can see ghosts, as the people of
St Kilda used to believe {Martin's St Kildal

A ghost-seer is rare in Australian tribes ; and the Semangs,
according to Mx, Skeat, think but Uttle of phantasms of the dead.
It almost seems as if some savages left ghosts behind, and applied
the animistic theory chiefly as a philosophical hjrpothesis. The
Arunta are notorious for their far-reaching and well-organized
philosophy of Animism, but seem not to see ghosts, or not often,
and are indifferent to the wants of their deceased and disembodied
friends. Other races, again, whose rehgion of ghost-feeding and
ghost-worship is based on animism, do not use it with the freedom
of the Arunta in their metaphysics and philosophy.

The more Animism, the less 'All Fatherism', if I may use Mr,
Hewitt's term for the superior being, such as Baiame, believed in by
many tribes in Australia and elsewhere. In Mr* Tylor *s theory of
rehgion as based on Animism this kind of being has his place, but
often there is nothing animistic in the native conception of his
nature. The opposite opinion has been caused by the loose employ-
ment of the word ' spirit', * gi*eat spirit', by European observers. In
his expected book Mr. Tylor may perhaps again consider this fact of
non-animistic religion. Meanwhile he began by exposing the vulgar^
error of denying to many races any vestige of religion ; an error
caused by narrow definitions of the term. As ^ a minimum definition
of Religion', he gives Hhe behef in Spiritual Beings'. Nobody can
define * Religion * so as to satisfy every one, and conceivably an

EDWAED BURNETT TYLOK

irreligious luind may believe in spiritual beings, while a religious
man may feel tliat he owes moral duties to a being whom he does
not envisage as a spirit. It is clearly * an idolon of the cave ', or study,
to regard such a belief as an advanced idea, beyond the reach of a
low savage. We have no notions of religion which low savages have
not developed, in their rough way, upon which we merely refine.
As Mr. Tylor says, * conceptions originating under mde and primitive
conditions of human thought, and passing thence into the range of
higher culture, may suffer in the coui-se of ages the most various
fates, to be expanded, elaborated, transformed, or abandoned. Yet
the philosophy of modern ages still to a remarkable degree follows
the primitive courses of savage thought, even as the highways of our
land so often follow the unchanging tracks of barbaric roads/

The most marked diflFerence between the third (1891) and the
fii-st edition of Primitive Culture was an extension of the theory of
savage and barbaric boiTOwings from the reUgions of Christianity
and Islam, But the material obstacle of stereotyped plates prevented
the author from working out this idea in sufficient detail. Mr. Tylor
expressed his views more fully in * The Limits of Savage Eeligion '
{J.A.L, 1891, xxi. 283).

* Timidly and circumspectly,' in his own words, Mr. Tylor has
sketched the outlines of his great system of the evolution of rehgion.
That kingdom cannot be taken by violence ; no fragile ladders of
hypothesis raised upon hypothesis can enable us to scale the^am-
mantia moenia.

On re-perusing the long famihar pages of Primitive Culture one
is constantly impressed anew by their readableness. Never sinking
to the popular, Mr. Tylor never ceases to be interesting, so vast and
varied are his stores of learning, so abundant his wealth of apposite
and accurate illustration. Ten years was this work in the writmg^
and it may be said that k temps fCy mord; that though much has been
learned in the last thirty years, no book can ever supersede Primitive
Culture, It teaches us that, in examining the strangest institutions
and beliefs, we are not condemned a ehercJier ration oil H n*y en apas^ as
Dr. Johnson supposed. The most irrational-seeming customs were
the product of reason like our own, working on materials imperfectly
apprehended, and under stress of needs which it is our business to
discover, though they have faded from the memories of the advanced
savages of to-day. We must ever make allowance for the savage
habit of pusliing ideas to their logical conclusions, a habit which our

EDWARD BURNETT TYLOR

English characteristics make us find it difficult to understand. We
are also made to see that man is, and will continue to be, a religious
animal. As Dean Swift acutely obseo^ed, * the Abolishment of the
Cliristian Religion will be the readiest Course we can take to introduce
Popery • . , and supposing Christianity to be extinguished, the
People will never be at Ease till they find out some other Method of
Woi-ship, which wiU as infaUibly produce Superstition, as this will
end in Popeiy/

Mankind, deprived of religion, would begin again at the
beginning,

For ghosts will walk and in their train,
Bring old rehgion back again.

While Primitive Culture is the basis of * Mr, Tylor's Science ', as
Mr. Max MUller called it, he has made many other valuable additions
to knowledge. Among these are his contributions to catecliisms for
field antliropologists. Many intelligent European and American
observers, among savages, are interested in, and desire to record,
what they see and hear, but are not acquainted with what is already
known to specialists, and are painfully vague in their terminology.
For their edification Mr. Tylor has drawn up eighteen sections in
* Notes and Queries on Anthropology' (British Association, 1874,
1892). In the Encyclopaedia Britannica (ninth edition) he wrote on
Anthropology, Oaths, Ordeals, Magic, Cannibalism, Divination, and
other topics. In the Game of PatoUi {J. A, Z, viii, 116) he investi-
gated the difficult theme of the diffusion, or independent invention
of this game in Asia and America or to America from Asia

It seems, at present, almost impossible to limit the range of
coincidence in invention, but this example stretched our ideas of its
powers to the uttermost. It is much to be desired that Mr. Tylor s
scattered contributions should be collected in a volume. His Gifford
Lectures at Aberdeen, still unpublished, form, it is to be presumed,
the germ of the great work with which he is still occupied. Since
1891 an enormous quantity of fresh information as to the customs,
institutions, and beliefs of backward races has come to our knowledge.
From Australia, Africa, and America we have received records, often
most carefully made, thanks to Mr. Howitt, Messrs. Spencer and
Gillen, Mrs. Langloh Parker; the students of the Bureau of Ethnology
at Washington, Mr, Hill Tout, and many other intelligent witnesses.
New hypotheses are not less conmaon than new facts, and the anthro-
pological world eagerly awaits Mr. Tyler's treatment of the evidence,

EDWAED BURNETT TYLOR

and liis criticism, if lie chooses to offer it, of the new theories. He
has never been a man of controversy ; his discussion with Mr, Herbert
Spencer {Mind^ 1877) had a foredoomed end* With all respect to
Mr. Spencer, he took up anthropology as a wdpepyov ; he was less
familiar with facts than fertile in conjectures, and much of his
reading was done by proxy, an impossible method*

Mr, Tylor has done one piece of vulgamat4on in his ' Antliro-
pology ', a manual of the subject, * an introduction * to the science.
Such manuals cannot ' go rather deeply ' into any pointy and I burn
to discuss with him his notions of the evolution of the shield, the
parrj'ing buckler and the great screen-shield. But, even on this
point, much information has accrued, which was not accessible to
any inquirer in 1881. The book contains ideas about the family
which are not fashionable, though I believe them to be correct. Mr.
Tylor 's theory does not start from the hypothetical promiscuous
horde, or Mn Howitt s * Undivided Commune *, and assume that the
family was slowly evolved out of that prinm 7fmteri€S, On the family,
he says, * the whole framework of society is founded/ wherefore the
family must be prior to the totem kin and the tribe, * Among the
rudest clans , , . the family tie of sympathy and common interest
is alread}' formed. • . .' Again, ' the natural way in which a tribe is
formed is from a family or group, which in time increases and divides
into many households, stUl recognizing one another as kindred. , . .'

Probably Mr. Tylor would now modify these statements, but it
does not seem probable that he will ever appear as the advocate of a
primal state of promiscuity. Whatever theory he may produce is
ceiiain to deserve the most respectful attention, for his combination
of wide knowledge and of sagacious caution gives his opinion an
unequalled weight in his own science.

It has been no part of my conception of my task to enter
into the details of Mr. lyior's biography. We know that, like the
minstrel of Odysseus, he was * self-taught ' in so far as he was the
alumnus of no University. In his youth the curricoolum (as the Scottish
baronet styled it) of the Universities did not embrace the study of
savtigery and of the advance from savagery. His example and energy,
with the munificent gift of General PittrRivers, the Museum over
which Mr, Tylor presided, have founded, in Oxford, a school of
Anthropology, though, as the undergraduate observed, * there is no
money in it/ Ours is a purely disinterested science. How much
Cambridge has done, and is doing, for Antliiopology, is known to the

EDWARD BURNETT TYLOR 15

learned world. Though we do not dwell on Mr. Tyler's biography,
we may regard him as a man not less serviceable than happy. His
genius has been favoured by the gift of leisure, and (may I be
permitted to say ?) by the long companionship of the lady who shares
his interests and aids his researches.

As to Mr. Tyler's poetical productions, their extent and merit,
his modesty forbids inquiry. I only know that, nee cUhara carens,
in collaboration with my weaker muse, he is the author of The Double
Ballade of Primitive Man.^ It was at first a single ballade of three
stanzas. Mr. Tylor^s additions raised it to the estate of a Double
Ballade oi ^\jL.

> In Ballades in Blue China.

DISTRIBUTION OF THE ORIENTAL FIRE-PISTON

Sc^/b for Afa/9

»o ao So ^ m
Scaim for Mans ® (3) ®

En^/ Afr/€« KX W vy

11.Y.4i<uJUsWiu^^ C»^A^ Hy.

THE FIRE-^PISTON

BY HENRY BALFOUR, M.A.
Curator of the Pnr-RivERa Museum, Oxford

The fire-piston appears to have been but little known to
ethnographers at the time when Dn E, B. Tyler published his
BeseaftJies into tJw Early Sistory of Mankind^^ which contains the
classical and fascinating chapter upon fire-making, one of the
pioneering articles upon this interesting subject. Dr. Tylor refers
(p. 246) very briefly to this instrument as follows : — * There is a well-
known scientific toy made to show that heat is generated by
compression of air. It consists of a brass tube closed at one end,
into which a packed piston is sharply forced down, tlius igniting
a piece of tinder within the tube. It is curious to find an apparatus
on this principle (made in hard wood, ivory, &c.) used as a practical
means of making fire in Bimiah, and even among the Malays,' If,
taking this short sentence as my text, I make an attempt to bring
together the available information regarding tliis peculiar fire-
producing appUance, I trust that I may, however inadequately, be
o£fering as my contribution to this volume a subject which at least
has the sympathy of the honoured and veteran anthropologist, to
whom the book is dedicated. Dr. Tyler's reference to the fire-piston
contains two statements, (1) tliat it is a well-known scientific toy,
(2) that it is a useful appliance in certain Eastern legions. I may
conveniently divide my subject in a similar fashion and deal firstly
with the * seientific toy * and its practical descendants as they exist
or have existed in civilized Europe, and secondly with the * useful
appliance * as it is found amid an environment of lower culture in the
East. An interesting ethnological problem is involved, one whose
solution is somewhat baffling.

Tlie Fire-piston in Europe.

Appreciation by physicists of the scientific fact that heat and
cold may be produced by the mechanical condensation and rarefaction
of gases, dates back to before the commencement of last century.

* London, 1878.

TTLOB C

THE FIRE-PISTON

A paper upon this subject was read by John Dalton in the year
1800/ giving the results of experiments in the compression and
rarefaction of air, which were noted as producing increased and
decreased temperatures. On December 29, 1802, M. Moliet,
Professor of Physics in the Central School at Lyons, announced
to the Institute of Prance that he had noticed that tinder could be
ignited by placing a small piece in the narrow channel with wliich
the lower end of a pump for condensing the air in an ordinary
condensation pump is furnished. Two or three strokes of the piston
were usually sufficient to cause a spark;^ He also stated that he had
observed a luminous appearance caused by the discharge from an
air-gun in the dai'k. On the strength of this announcement,
J. C. Poggendorff^ refers to Mollet as the discoverer of the
Tachypynon (instrument for producing fire by compression of air).
On the other hand, we may gather from F. Rosenberger* that a
workman in the small-amis factory at Etienne-en*Forez (near Lyons)
was the actual discoverer of the fact that a great amount of heat was
generated in charging an air-gun with an ordinary compression-
pump, and that tinder could be ignited thereby. Mollet is here
stated to have communicated this discovery by the workman, who
must, if Rosenberger's account is the true one, be credited as being
the original French observer of this phenomenon, Mollet having acted
as the reporter of the discovery. The facts announced were not
understood by the French scientists, who were inclined to discredit
them, but very soon the experiment with the air-compression pump
was repeated by others, and tinder {amadou} was easily ignited by
this means. A letter was sent by M, A. Pictet, one of the editors
of the BibUctheque Brifannique^ to Mr, Tilloch in England, on
January 1, 1803, announcing Mollefs communication to the
Institute of France,^ and the writer stated that he considered the
phenomenon as never having been noticed before. But WiMiam
Nicholson affirmed ^ that it (the flash from an air-gun) had been

^ Mem. Mafichester Lit* and FhiL Soc^ v, pt. ii, p. 515, 1802.
' Journal de Physique^ Iviii, 1804, p. 457; Nicholson's Journal of Nai, PhUo-
scjph^i Chemistry, and the Aris^ iv, 1808 ; Philosophical Ma§Qsine, xiv, p. 363,
^ Biograph-Literarisches Mandwcrterbuch, ii, 1863, Leipzig.

* Geschichte d. Phtjsih 1887, iii, p. 224.

* Phiiosophiml Magazine, xiv, p. 363.

* Nkholson*s Journal, 1 c. ; Marc Auguste Pictet, * Sur r^chanffement dea pro-
jectiles par leur frottement contre rair,' BibUotheque Britanniquej xxiii, 1803,
pp. 331-6.

THE FIRE-PISTON

known for some time in England, having been first mentioned nearly
a year and a half previously by Mr. Fletcher at a meeting for
philosophical experiments and conversations, which was then held
weekly at Mr. Nicholson's liouse. He adds, * It is a curious
phenomenon, and deserves investigation.' No one at the time
explained the cause of the phenomenon, which had been accidentally
noticed and had not been arrived at by direct scientific experiment.
Nicholson's statement is interesting, not only as assigning the fii'st
observation of this physical effect to an earlier date {somewhere
about the middle of the year 1801), but also as ascribing to an
EngUslmian its discovery.

In later days the experiment of igniting tinder in a compression-
pump became a conmion one in physical laboratories, and fii-e-pistons
were specially made for the purpose. Tliese consist usually of
cylinders of brass, closed at the lower end and very accurately bored
or gauged. Into the bore fits a piston-rod, carefully packed at the
lower end, so as to occupy the bore os completely as possible. At
the lower exti'emity of tliis piston-rod is a cup-like depression, in
which a piece of ammlou can be placed. By driving the piston-rod
home very forcibly the column of air in the cylinder is violently
compressed into a fraction of its normal length, the sudden conden-
sation generating an amount of heat amply sufficient to ignite the
tinder. The piston-rod is at once withdrawn as quickly as possible
and the tinder is found to be glowing, and a sulphured match may
be lighted from It. In place of the brass cylinder and piston one of
glass may be used, and the vapour of carbon bisulphide can be
exploded by the compression, the flash being plainly visible through
the glass.

Not only was this principle adapted for scientific illustration,
but it was also applied to domestic use. Wlio was the first pei-son
to adapt the air-compression method for use in everyday life may
never be known. We know, however, that its potentialities for
utihtarian puiposes were recognized not very long after the scientific
interest had been roused. Among the specifications of English
patents for the year 1807, there is one, dated February 5, number
3007, recording an invention by Richard Lorentz of * an instrument
for producing instantaneous fire '. The figure of this instrmnent is
reproduced here (fig, 1), and the specification runs as follows : —

'Tlie illustration shows the construction of my machine or
instrument for producing instantaneous fire, a represents the cap

THE FUtE-PISTON

or head of a staff or stick, having therein a cavity or space for
containing tlie prepared fungus known by the name of German
tinder, or for containing common tinder of rags, or any other very
combustible substance, c is the outer end of the rod of a syringe,
which works by a piston in the upper part of the staflF, and by a
stroke of about twelve inches forces the common air with great
velocity and in an highly condensed state through a small aperture
against the combustible matter included in the head a, which is well
screwed on against a shoulder or face armed with a collar of leathers.
h is the hole for admitting common air when the piston is drawn
quite back. The manner of working consists simply in pressing the
end of the rod of the charged syringe strongly against the ground
so as to drive the air suddenly on the tinder, and the cap a being
without loss of time unscrew^ the tinder is found to be on fire/

It will be noted that this instrument differs in one important
I>articular from the ordinary fire-piston of the physical laboratory.
In the latter the air is merely compressed in the bottom of the
cylinder, whereas in Lorentz's machine the air is not only com-
pressed by the drive of the pis1;on-rodj but it is also forced under high
pressure through a minute duct beyond which the tinder Mes. The
term fire'Syrhige^ so frequently applied to the various instruments for
producing fire by air-compression, seems to be peculiarly suited to
tins form, since the air is forced through a duct at the end of the
main cylinder-chambery just as water is forced through the nozzle-
duct of a squirt or syringe. No doubt the air, already heated by the
compression, gains additional heat from the friction caused by its
violent passage through the small duct. It is possible that tliis
instrument owes its origin to the observation of the flash produced
by the dischai'ge of an air-gun, to which I have referred above, in
addition to the scientific experiments as to heat generated by simple
compression of air in a small space.

Fire^pistons in wliich the duct was omitted appear to have
enjoyed some favour upon the Continent, and to a lesser degree in
England, during the early third of the nineteenth century. In the
Mechanics* Magazine^ voL xvii, 1832, p. 328, the following passage
occum : —

* The following is a sketch of a simple instrument for obtaimng
a light. As the invention though not new [It is very well known
iiii the Continent by the name of the Instantaneous Light-giving
Syringe, As it has not, however, been described in this work and
may be new to some of onr readers, we insert our correspondent's
description, — Edilorial note] is, perhaps, not generally known, I shall

THE FIRE-PISTON

be glad to see this description of it in your magazine . . . yours
respectfully, R J. Mitchell, June 19, 1832/

The description referring to the figures (fig. 2) I give in full.
"ab is a brass cylinder^ similar in appeaiance to a small brass
cannon, having the hole rather better than | of an inch in diameter,
drilled true and clean rinsed. CD is the form of a piston to work in
the cylinder, but unpacked, bf is the same ready packed with
tliick leather and fitted up for use, h is a circular brass nut,
working against the screw to keep the packing tight, k is a small
hook, fastened in a hole drilled through the nut h, c is the handle
to the piston and is made of wood. The method of use is described
as follows. * Prepare some thin cotton rag (older and thinner the
better) by steeping it in a solution of saltpetre, and drying it in
a warm oven ; tear a small piece off and place it on the hook k ;
introduce the piston ef into the cylinder ab a short distance only ;
then take the cylinder in the right (sic) hand, place it perpendicular
upon the floor or a table, and strike the handle e with the ball of
the right hand, so that the piston may rapidly descend to the bottom
of AB, and being mddenly withdrawn, the tinder will be found on
fire, and will light a common brimstone match. AtnudoUy or German
tinder, which may be obtained at any of the principal druggists, is
likewise a good tinder, but I prefer the rag steeped in saltpetre,
E. J. M,' This instrument is of the simple air-compression kind,
and, except for the piston-rod terminating in a hook instead of a
hollow for the tinder, it is identical in principle with the most
prevalent form of fire-piston.

In 1834 a notice occurs ^ of a French modification (fig. 3) of the
type of instrument invented by Lorentz, referred to above, though
from the following accoimt it does not seem to have been very
successful : —

*An attempt has been made in France to produce an instan-
taneous light by the compression of air, A strong tube a is
furnished with a piston b, wliich may be driven rapidly from c to
D by striking the knob e at the end of the piston-rod. The end of
the tube, at d, is pierced with small holes to allow the air, when
forced up by the piston, to pass into the hollow space o, in the piece
p, screwed air-tight to the end of the tubes. When a light is
wanted a bit of tinder is placed in the hollow, the top screwed on,
and the piston driven in forcibly ; on unscrewing the top the tinder

> The Penny Magazine, London, 1884, July 26, p. 286.

THE WtSM^FBttm

tiiB timfer ham
the

will he fotmd

getting a light, raqpMs # flMtah to te
taken ftre, aad k

Theowllbdiif .
a liiiiTu iliiinlww i» ^^7 obbSmt to ifaat
Io90 of time eammi by the
raeqitoefe after the tindtr m
file ^Adeoer of theM 4^^9940:11 for^ As ikr as I am «mre imbd
ef Asm luvis bMA |Himi lid , SDd this maybe ao imitiieatMP Aat
th^ neyer were nimieiuiii or extoaiptlf l a wd, Spseneiis of the
simple fira^nslmi form oeeor wyaiingly in mnseoma aod prrrate
coUeetionsk An eyampfc^ from Bedmioalsf:^ Bmsks CouBtr, PennsrU
vania, said lo date about 1815-20, i^ oMi^iimed bj H. CL MCToer.^
A specimsii of biast from Gestrikland, or Hebinglm d, Sweden, is
in the Nordislui Mnseom, Stoekholm. Mr, IL Bidwell po otHMOQ three
specimens, one of whkh (fig. 4) £9 entirelj of brass and of lafge siae,
and resembles rather the moduli instrament of the phyaieal
laboratoiy than the old domestie form. The tube is of thin bn^s,
S| inches long by | inch in diameter. Hie other two (figs. 6 and 6),
which msLy have been intended for domestic nse, are smaller, of
brass throughout, with the exception of the piston*rods, which are of
uteeL In one (fig. 5) the lower end of the piston-rod is packed with
leather, while in the other (fig. 6) a brass piece accurately fitting the
bore of the tube is screwed on, and no packing is required. All
these have cup-like depressions at the end of the plunger for holding
the tinder. Mr. BidwelFs specimens are all said to be English. It does
not appear likely that the practical everyday use of these fire-pistons
was at any time very genei*al, and the tinder-box eaaOy held its own
against them, but it is worthy of note that a certain practical value
was recognized for them, and even in quite recent years they were
reintroduced in France, and a pocket form was sold by tobacconists
in Paris. In these (fig. 7) the cylinder is of white metal with wooden
knob, the plunger is of hard wood with cupped end, and fixed to the
side of the cylinder is a tubulai- holder for the common cord tinder,
A specimen given me by Mr. Miller Christy w^orks very satisfactorily
with a reully ' quick* form of tinder. Its reintroduction in Western
liuru|>o was, no doubt, prompted rather by its peculiar interest as a

^ Light and Fire-maMnfff Piiiladelphi&, 1896, p. 26.

THE FIRE-PISTON

scientific toy than by its being recognized as being of real practical
importance. For ordinary purposes^ as an appliance useful in
everyday life, its death knell was sounded when the lucifer match
became generally known. The latter, which has held its own un-
challenged during the last seventy years or so, proved too strong and
too severely practical a competitor^ before which the time-honoured
tinder-box, the fire-piston, and the earUer chemical methods
(* sulphuric-acid bottle/ * phosphorus bottle,' ' promethian,' &c.)
had to give way*

The Fire-piston in the East.

Interesting as is this fire-producmg appliance as it occurs in
Western Europe in the form of a scientific instrument, and, to a
limit extent, as a machine for domestic use, from an ethnohgical
point of view, the interest of the fire-piston centres mainly upon its
occurrence in the East in an environment of relatively low culture.
The problem is to ascertain whether this peculiar and very specialized
method of fire-production was introduced into the Oriental regions
from Europe, or whether it was invented independently by the little-
civilized peoples among whom it is found as an appUance of practical
ever}'day use. Either theory is besot with diflSculties, which are
likely to remain unsolved in the absence of early records. 1 shall
revert later to the consideration of this question, and will now deal
with the geographical distribution and varieties of the fire-piston in
Oriental regions. Briefly stated, it may be said that the range of
this instrument extends sporadically over a wide area from Northern
Burma and Siam through the Malay Peninsula and the Malayan
Archipelago to its eastern hmits in the Islands of Luzon and
Mindanao in the Philippines,

Burma. — In this region the fire-piston is principally associated
with the Kachin (Kachyen, Kakliyen, or Kakyen) people, and the
forms vary as regards the materials used in their construction. The
cylinders may be of bamboo, wood, or horn, the pistons or * plungers '
are either of wood or horn, or are made of a combination of both
materials. In all, the heat is produced by simple compression of
the air in the tube, and I have seen no examples in which the air is
forced through a duct.

Four examples were collected for me by my friend Mr. Et, E.
Leveson from the Kachins, on the Cliinese border of the Northern

THE FIRE'PISTON

Shan States (Lat 24' T N., Lon. 98^* 15' E.), nearly due easi
Bhamo. These are interesting on account of their rude and simple
structure. Each (figs. 8, 9, 10) consists of a natural tube of stout-
walled bamboo, closed near the lower end by a natural node. The
* plunger ' is of wood, with large roughly-shaped head. One of the
heads is hexagonal, each facet being decorated with chip-carving
(fig, 10). The lower end is cupped to form a receptacle for the
tinder, and is packed with fine thread coated with wax (?). Two
very similar specimens from the Shans of Upper Burma are in the
Ethnological Museum at Cambridge.

A better made example, though stUl composed of the same
materials (fig. 11), was collected for me by Mr. Leveson from the
Wa villagers in East Manglun {Mong-lem), on the Chinese frontier,
22° 20' N,, 99" 10' E, The bamboo tube is neatly finished off, and
the * plunger ' is of very hard wood, with exceptionally large head
accurately shaped. Another specimen in my possession (fig, 12)
having a cylinder of bamboo is somewhat more pretentious, the
eyUnder being carved in a decorative form ; the wooden ' plunger '
is unusually long and tapering. This example was obtained by Mr.
Frank Atlay at the Ruby Mines, Mogok, and kindly given to me.
A small cloth bag containing vegetable-floss tinder belongs to this
specimen, with which I have been able to produce fire with
considerable ease on many occasions.

In the Ethnological Museum at Rome are several very rudely
constructed examples of wood and horn, collected by Leonardo Fea
from the Kachins (Cowlie Kachins) and Shans in the neighbourhood
of Bhamo, chiefly to the east of that town. These (figs. 13, 14, 15,
16) differ in some respects from the types most commonly seen from
Burma. In all of them the cylinder is of stout buffalo-horn, either
light or dark coloured, cut from the solid tip of the horn. In two
of them (fig, 13) a pair of flanges are raised upon the surface near the
top, and a carrying cord is knotted through these flanges. A similar
pair of perforated flanges appear on a specimen in the British Museum.
In these two examples the * plunger ' is of hard wood, with expanded
head cut from the solid (fig, 13). A third specimen has a piston,
with wide head cut from one soHd piece of dark horn. Three others
(figs. 14, 16, 16) and a fourth specimen (fig, 17) from the same
district, given me by Professor E, H, Giglioli, are peculiar in having
the shaft of the plunger of horn, while the head is of wood fixed by
means of a stout rivet of horn to the shaft, which is vridened at this

I
I
I

I
I

THE FIRE-PISTON

point, and is tenoned into the head. The head in some consists
of a single piece of wood, in others it is in two pieces, and is
reinforced with bindings of string and cane. The riveted head
seems to be specially associated with the Kachins^ The collector
gives the native name of the instrument as caifo or caifoe^ and he
adds the remark that while these people are called Kachin by the
Burmese, they describe themselves as Chimfo or Simpfo (i* e. * men *) ;
the name is also given as Chingpaw,*

A specimen (fig, 18) in my collection, obtained by Mr* Leveson
from a Kachin on the Chinese border, from the same district whence
the ruder bamboo specimens were procured, has a cylinder of rough
horn of a light colour and a plunger, also very- roughly made of black
horn, Keference is also made by Capt. W, Gill ^ to the fire-piston
(with wooden cylinder) amongst the Kachins of the village of Pung-shi
(Ponsee), on the Taiping Eiver, fifty or sixty miles east of Bhamo.
John Anderson^ describes and figures the instrument from the
Kachins of the same region ; it resembles that shown in fig. 18.

Other specimens of the Kachin fire-piston of which I have
record are as follows : —

2 examples with plain horn cylinders, Berlin Museum.

1 (referred to above) with horn cylinder, 8-7 cm. long, having

perforated flanges for a carrying cord ; * plunger * of hard

wood riveted to rounded wooden head ; given by Mr. R,

Gordon to the British Museum in 1873.
1 given by Mr. R Gordon to the Mayer collection, Liverpool

Museum, 1874.
1 of wood, Horniman Museum.
1 with tapering cylinder of horn and wooden * plunger', in

Mr. E. Bidwell*s collection.
1 with tapeiing horn cylinder, 7*5 cm. long, piston of horn

tenoned into a cubical wooden head and secured with a

rivet ; given by Sir W. N. Geary to the British Museum,

1901

' H. J, Wehrli, IntemaL Archiv f, Ethnographie, suppl. to voL xvi, 1004, p* 45.
Pfiee also L. Fea, QucUtro Anni fra i Birmanni e le tribit limitro/e, and E. C. J. George,
Ijfenioirs on the Tribes inhabiting the Kachin MillSf Census of 1892, Burma Heport, i,
fipendlces.

' The Miver of Goldm Sand, 1880, vol. ii, p. 395.

' Mandahy to Momien^ 1876^ p. 184, and plate, figs. 3 and 4.

m ^^ THE FIRE-PISTON

1 with cylinder of horn, 8-6 cm. long, tapering upwards, cut in
nine longitudinal facets, and mtli ring of cai^ving round the
base ; plunger of hard wood with the head capped with silk
wrapping; native name, mi-put; given by Captain R C*
Temple to the Pitt-Rivers Museum, Oxford, 1890 (fig. 19).

1 small though elaborate specimen of black horn throughout,
apparently lathe-turned, the cylinder ornamentally shaped,
and reinforced at the end with metal bands, as is also the
rounded head of the ' plunger ' ; from the cylinder hang
three strings, one carrying a small velvet bag of vegetable-
floss tinder, another a small nut-shell containing grease for
lubricating the packing of the piston, the third a small
ivory spatula for spreading the grease {fig. 20) ; given by
Major R, C. Temple to the Pitt-Rivers Museum, 189i.

From Mandalay, and probably of Kachin origin, I have in my
possession a specimen very similar to the last, of black horn
throughout, lathe-turned, the head of the piston riveted to the end
of the shaft ; with bag of vegetable-floss tinder, and small, spherical
wooden grease-box (fig. 21) ; given to me by the collector, Mr. H. 0.
Mordaunt, in 1899.

A sketch (fig. 22) of an elaborately car\^ed fire-piston seen in
Mogok, 1893, was made for mo by Mr. Donald Gunn. The decorative
treatment of this specimen is unusually elaborate. The native name
is given as mizomh

Two examples, locally called mi-put^ collected in the Southern
Shan States, were given me by Mr. H. E. Leveson in 1890 and 1891.
Of these, one (fig. 23) is quite plain, with long cylinder of hard
wood, and piston of buflfalo horn with large rounded knob. It was
obtained from a ptifigi^ or priest in a monastery (kyaung). The other
(fig. 24) is entirely of buffalo horn, the cylinder gracefully fluted in
eight facets ; the plunger is elegantly tapered, and has a rounded
head inlaid with small metal studs. The depth of the bore in the
cylinder is only 4 5 cm., the cylinder itself being 88 cm. long. This
gives a very limited play to the piston, rendering the operation of
fire-producing a somewhat difficult one. Belonging to this specimen
are a tinder-pouch of pahn-spathe and a turned-wood box for grease
(fig. 24 a).

Further still towards the south a specimen was seen by Prof.
A. Bastian, in a monastery in Shwegyin, which lies near the mouth

THE FIRE-PISTON

of the Poung-loung River in Pegu. The tube was of ivoryj A
similar specimen was made for him by a native.

It would appear that the westerly limit of distribution of fii-e-
pistons in Burma is bounded by the Irrawaddy River, while they extend
in a north and south direction from the neighbourhood of Bhamo to
Rangoon. To the north-east they extend some distance acroas the
Chinese frontier^ amongst the eastern Kachins and peoples of mixed
Kachin blood. On the eastern side of Burma they are found in both
the Northern and Southern Shan States,

French Indo-China, A fire-piston (fig, 25) in the Edinburgh
Museum was obtained from the Khas (or Kumuks), an aboriginal hill
tribe of low stature, inhabiting the country north of Luang Frabang,
which lies on the Mekong River in Lat. 20'* N, It is entirely of horn :
the cylinder is carved in an ornamental moulding at either end.
The piston has a knobbed head coated, appaiently, with some kind
of composition. A bag of cloth serves as a tinder-pouch (fig. 25 a).

Further to the south-east, the implement is again met with
amongst the Mois^ a people of very low culture inhabiting the table-
lands and mountains between the Mekong River and the coast of
An nam, ftom the frontier of Yunnan to Cochin China. They differ
racially from the Annamese and Thai, and are said by Deniker ^ to
belong probably to the * Indonesian ' stock. A. Gautier describes ^ the
instrument as having a cyUnder of hard wood, with a bore of 7 to
8 cm. in depth, and 7 to 8 mm. in diameter. The piston, also of wood,
has a large, rounded knob, and is cupped at the lower end for the tinder
in the usual way. The tinder {amadou} is kept in a hard fruit-shell
hollowed out. The native moistens the end of the piston in his
mouth, so as to lubricate it, and also to make the small piece of
tinder adhere to the cupped hollow. Apparently the instrument is
in constant use amongst the Mois.

Malay Peninsula. John Cameron frequently saw the fire-piston
in use amongst the Malays of the Straits, prior to 1865. He writes,^
' There is one peculiarity which I will mention, as it might, I think,

* Bnatiftiif Die Voelker des Oesilichen Asiefif 1866, ii p. 418.
' Maces of Man, p. S92.
' • 6tude BUT lea Mots,' BuU. (k la SocUt^ de G^raphk Coninicrcialc du Havre,

1902, pp. 95 and 177.

* Our Tro^^icaX Possesmns in Mo^atfan India^ 1865, p. 136.

THE FIRE-PISTON

be capable of improved application at home ; it is the method
adopted by some of obtaining fire. It is true that this is not the
usual method, nor do I remember to have seen it alluded to by any
other writer ; I have witnessed it, nevertheless, repeatedly availed of
by the Malays of the Straits ; and in some of the islands to the east-
ward of Java, where I first saw it, it is in constant use. A small
piece of horn or hard wood about three or four inches long, and
three-quarters of an inch in diameter, is carefully bored through the
centre for three-fourtlis of its length, with a hole about a quarter of
an inch in diameter. To fit this, a sort of ramrod or piston of hard
wood is made, loose all along, but padded with thread or cotton at
the point, so as to be as nearly air-tight as possible, when placed into
the hole of tlie little cylinder. . . . When used, the cyUnder is held
firmly in the fist of the left hand ; a small piece of tinder, generally
dried fungus, is placed in a cavity on the point of the piston, which
is then just entered into the mouth of the bore ; with a sudden
stroke of the right hand the piston is forced up the bore, from which
it rebounds sUghtly back with the elasticity of the compressed air,
and on being plucked out, which it must be instantly, the tinder is
found to be hghted, . , , I can only attribute the light produced to
the sudden and powerful compression of the air in the bore of the
cylinder/

This description of the method of using the fire-piston appUes
practically to aU Oriental examples. The record is interesting as
being an early reference to the use of the instrument in the Peninsula,
and also in the Eastern Malayan Archipelago.

Turning now to more recent records of the occurrence of fire-
pistons in the Peninsula, I may give the following first-hand informa-
tion, which I owe largely to Mr, W. W. Skeat and to Mr. Nelson
Annandale, who have done so much for the ethnology of this
region.

Mr. Annandale, in 1901, saw the instrument in regular use at
and in the neighbourhood of Ban Sai Kau, a village in the State of
Nawnchik (called Toyan by the Malays), the most northerly of the
Siamese Malay States, west of the Patani River. The Siamese name
of the fire-piston is lek phm toK\ the Malayan name is gobi apL It is
there chiefly used for lighting cigarettes in the jungle, as the spark
is not easily extinguished by high winds. One specimen from this
village, given me by Mr. Annandale (fig. 26), is of very small size, the
cylinder being only 6'7 cm. in length and the bore 4 6 cm. It is

THE FIRE-PISTON

entirely of black horn ; the cylinder is ornamentally, though rouglily,
tumedj barrel-shaped in the centre, and tapered to a blunt point at
the lower extremity. The piston has a plain, rounded knob at the
top, and the usual hollow for tinder at the other end, A specimen
obtained there by Mr, Robinson for 5 cents is very similar in shape
and size, though somewhat better made. A third specimen from the
same locality (fig* 27), collected by Mr. Annandale for the Pitt-Rivera
Museum, has a very elegantly lathe-turned and slightly engraved
cyUnder of horn ; the piston is of light wood with a turned knob of
horn through which it is fixed with an adhesive.

From further south, in the State of Patani, Mr, Skeat procured
three examples very similar in shap>e to those of Nawnchik, these
are in the Cambridge Museum. One of them is very small (fig. 28),
with horn cylinder and wooden piston ; the depth of the bore is
only 3 COL A second has a lathe-turned horn cylinder and a piston
of hard wood with ivory head, depth of bore 3-7 cm. The third
(fig, 29) is larger somewhat, with lathe-turned cylinder of bone and
wooden piston; depth of bore 5*5 cm. All three were obtained in
Jalor (Jala), one of the seven districts of Patani, some thirty miles
up the Patani River. The Malay name given by Mr. Skeat is gobek apt
(lit. * fire-piston'). The word gobek is that usually appUed to the
piston (pestle and mortar) used by old and toothless men for crushing
up the fceieWeaf ; api in Malayan means ' fire'. The tinder, rahok^ is
usually the fluflFy substance obtained from the leaf-bases of the tukas
palm {Caryota Griffithn}^ though occasionally it is obtained from other
kinds of palm, or from rattan. Mr. Skeat tells me that the fire-piston
occuiB throughout the interior of the old Malay state of Patani, in
other words, the sub- districts of Jala, Ligeh, Biserat, and Rliaman,
and he also mentions that there is a probable extension northward
and eastward into more distinctly Siamese territory. His specimens
are practically identical with those obtained by Messrs. Annandale
and Robinson in Nawnchik.

Mr. Annandale procured for the Pitt-Rivers Museum an
example from the Samsam (i.e. Siamesing-Malay) village of Ban
Phra Muang in Trang, on the west coast, c. 7" 25' N,, 99^ 30' E,
This is the most northerly district in the Peninsula from which
I have definite record of a fire-piston. This specimen (fig. 30) has
a cylinder of light-coloured horn, pointed and ringed below, as
usual in the Peninsula, the upper half roughly bound round with
string coated with black wax. The piston is of black horn mth

THE FffiE-PISTON

rounded^ carved knob, which is hollowed out as a receptacle for
holding the supply of tinder. The depth of bore is 5-5 cm.

There are specimens in the Taiping Museum from the province
of Perak, on the western side of the Peninsula, but their exact
locality is not specified, and I have no descriptions of them as yet.

An interesting aberrant type (fig, 31), now in the British
Museum, was sent to Mr. F. W. Rudler in 1893 by Mr* Henry
Louis. It was obtained by the latter in 1890 when in camp on a
little stream known as Ayer Katiah, a tributary to the Teluban River,
Presumably this is the Telubiu River in the Siamese States of Saiburi
or Telubin, the next river down the coast after the Patani River, In
this the cylinder is of wood, 6-4 cm, long, neatly bound round with
bands of plaited cane. The lower end is rounded off, instead of
terminating in the point so characteristic of the Peninsula* The
piston, of hard wood, is very short, and has a large, roughly-carved
head* The packing is of pale vegetable fibre. A large bean-shell
serves as a tinder-box : it appears to be an entada bean (fig. 31 a).
Mr. Louis related that a party of Malays came down from some
neighbouring kampongs {i.e. villages), and squatting down in camp,
began to smoke, when one of the party, a young man, in the most
matter-of-fact way, took out his fire-piston and lit his cigarette. The
particulars were kindly sent to me by Mr. Rudler,

It will be seen that the distribution of the fire-piston is a very
wide one in the Malay Peninsula, where it is found in the hands
of both Malay and Siamese people, as weU as among the mixed
Siamese-Malays. The question arises whether the instrument is
originally Malayan or Siamese. 1 have consulted Mr. Annandale
and Mr. Skeat upon this point, and both are inclined to regard it as
of Siamese origin. The former writes to me as follows : * With
regard to the gobi api^ it is, so far as I am aware, a purely Siamese
implement I have never seen or heai'd of it in a purely Malay
community. , . . There are specimens from Perak in the Taiping
Museum, but their exact locality is not recorded, and even within
a few miles of Taiping there is a large Samsam village, while the
people of Upper Perak are indistinguishable from those of Rhaman
and Kedah, being physically as much Siamese as Malay.' Mr. Skeat
infonns me that, although the specimens which he obtained in Jalor
were used by Malays, he is inclined to think that they are borrowed
from the Siamese (or Siamesing-Malays), who appear to use them
much more than the Malays do. * There are a good many Siamese

THE FIEE-PISTON

and Samsams (i.e. Sianiesing-Malays) in the district, and it is to
their influence that I am inclined to attribute these fire-utensils/
Again he writes, * I have a strong belief that this particular object is
Siamese, because it appeared to die out as we worked south into the
more exclusively Malay districts, and I never came across any
specimen of it in Kelantan or Trengganu (wliicli are substantially
Malay districts), any more than I did on the west coast, where
Siamese influence was equally at a discount My recollection is
quite clear on the point that at Biserat in Jalor the fire-piston was
used by the Siamese more commonly than by the Malays, who
appeared to have borrowed the idea from them/

I have not as yet seen or heard of any specimens of the fire-piston
from Siam proper, but it would be most interesting to know if they
have been used there, and also to learn the details of their form, so
that we may ascertain whether the types of the Burmese region
can be linked by intermediate varieties with those of the Malay
Peninsula,

I must now turn to the distribution of this interesting fire-
making appUance beyond the southern limits of the Peninsula.

Sumatra.. — Van Hasselt ' mentions the use of the firo-piston by
the Menangkabo Malays in the hinterland of Padang, on the west
pde of the island. The specimen which he describes (fig, 32) is of
' karbouw ' (buffalo) horn, and its native name is tjaioew apt hUlaniaq.
In form it reminds one of some Kachin types. Its size is large, and
the plain surfaces of both cylinder and piston-head are relieved vdth
ring marks. The tinder, rahoewq (cf. rahok in Jalor), is obtained from
the anau palm. This specimen was obtained at Soepajang.

There is a specimen in the Berlin Museum from Padang on the
west coast, but of this I have not full particulars,

Mr, R. T, Pritchett figures ^ an ornate example from Sumatra
(fig. 33) ; he does not, however, specify the material or the size.

There is a very fine specimen in the British Museum (fig, 34)
which was collected by Carl Bock at Fort van de Capelle, Padang
province, Sumatra. This example is elaborately cai*ved out of horn.
The cylinder is 8-2 cm, long and tapers slightly from above ; it is
decorated with bands of carving. The piston has a carved head
which is surmounted by a well-shaped, rounded receptacle for tinder.

' Veth, Midden Sumatra, iii, p. 177, and pi. Ixxxiii, figs, 12 aad 13.
* Smokiana, 1890, p. 97.

THE PIRE'PISTON

This is very neatly fitted with a cap or lid which fits into the
opening like a stopper, and is furnished on one side with a small
projecting spur ; in closing the lid this spur passes through a slot in
the rim of the tinder-receptacle, and a half-turn secures the lid in
position {fig. 34 a). The name of this instrument is given as t^inar
datar^ but it seems possible that there has been some confusion with
Tanah Datar, the name of a place. At least this name requires
verification.

Borneo — Sarawak,— In this island the fire-piston is found
principally in the hands of Malays and Sea Dayaks of Sarawak.
In 1865, F. Boyle described ^ it as used by some of the Dayak tribes,
and expressed much astonishment at the singular method of procur-
ing fire. His description is evidently erroneous, but he adds,
* I must observe that we never saw this singular method in use,
though the officers of the Rajah seemed acquainted with it.' He
refers to lead being used as a material in making the instruments,
and adds that * the natives say that no metal but lead will produce
the effect *.

Charles Brooke, in 1866,'^ writes as follows: * There is a method
. . . used by the Saribus and Sakarang Dyaks for obtaining fire,
which is peculiarly artistic, and from what direction such a practice
could have been inherited is beyond my ken. The instrument is
a small metal tube, about three inches long, closed at one end, with
a separate piston, the bottom of which fits closely into the tube, and
when some dried stuff answering the piu'pose of tinder is introduced,
and the piston slapped suddenly down, the head of it being held in the
palm of the hand in order to withdraw it as quickly as possible with a
jerk, fire is by this means communicated to the tinder in the tube*
The Dyaks call the instrument " besi api ". '

W- M. Crocker asserts ^ that the fire-piston is ' found amongst
the Saribus Dyaks only. Here we have a small brass tube lined
with lead : no other metal, the natives say, would produce the same
result. A small wooden plunger is made to fit the tube, the end of
which is hollowed out in the shape of a small cup, in which is placed
the tinder/

* Adventures among the Thfoks of Borneo, 1865, p. 67,

* Tmi Tears in Sarawak, 1866, p. 50.

■ Jourrh Anthrop, Inst., xv, 1886, p. 426,

THE FIRE-PISTON

W. H. Fumess also describes and figures ^ an example with
lead-lined brass cylinder and wooden piston, from the Saribas Ibans
(Sea Dayaks), and in the British Museum there are two specimens
from the Saribas district, also Sea Dayak. One was presented by
Mr. G* D. Haviland in 1894, the other by Mr. Charles Hose. Both
instruments have cyhnders of lead^lined brass, 9 cm, and 9 8 cm.
long, and pistons of hard wood. Mr. Hose's specimen has attached
to it a bamboo box for tinder, the other has a tinder-holder of
canarium nut-shell and also a small cleaning-rod of cane and a metal
spatula (? for grease). BeM api and gochoh api me given as the native
names. These two examples closely resemble a specimen (fig, 35)
presented by Mr. D, L S. Bailey to the Pitt-Rivers Museum in 1904,
It came from the Sea Dayaks of Simanggang, near Saribas. In
structure it is identical with the others, and it has a tinder-box of
canarium nut and a brass pricker attached to it. Dn A, C, Haddon
brought back a very similar Saribas Dayak specimen, gtichu api.
Another example of the same form in the Kuching Museum, said to
be from the Kayans but doubtless of Sea Dayak origin, is figured by
both Lady Brassey ^ and R T, Pritchett.^

Another type of fire-piston in Sai-awak differs from the above
only in the fact of the cylinder being made of lead alone, instead of
the lead being merely a lining to a brass tube. Mr. D. L S, Bailey
presented a specimen of this kind to the Pitt-Rivers Museum in 1904
(fig. 36), The cylinder has been cast evidently in a two-piece mould
of bamboo, and is composed of a mixture of lead and tin. It is
decorated with simple relief designs. The piston is of wood.
Attached to the cylinder are a tinder-box of entuda bean full of palm-
scurf tinder, and also a brass-wire pricker. It is a Sea Dayak
specimen from Simanggang,

A nearly identical specimen was given to the Pitt-Rivers
Museum in 1889 by Mr. S. B, J. Skertchley. It was made by a
Kalaka (? Kalukah) native from the western part of Sarawak, not
very far from the Saribas and Simanggang districts. Mr. Skertchley
gives a detailed account of the instrument, to which I will refer
readers for full details, and also an excellent figure/ The instrument
itself, best api^ resembles the last in all essential details ; a bamboo
tinder-box with palm-scurf tinder, a cleaning-rod of cane, and one

* Home Life of the Borneo Head-hunters, 1907. p. 170.

* The Last Voyage, 1887, p. 148. ' Smokiana, 1890, p, 97.

* Jmm» Anthrop. Inst,, xix, 1890, pp. 445-8, and pi xi, fig. 1,
TLom D

THE FIRE-PISTON

half of a bamboo casting-nioiild accompany the specimen. Mr.
Skertchley says that the metal of the cylinder is composed of two
parts lead to one of tin. * It is cast in a bamboo mould. . . . The
mould is a thin piece of bamboo, split lengthwise, on the interior of
which the ornamental bands, &c.j are incised, A piece of flat wood»
plank by preference, has a hole made in it the size of the bore.
Through this hole a rotan is pushed, which also passes through a
lump of clay tempered with sand stuck on the upper surface of the
plank. Tlie rotan projects beyond the clay to a distance somewhat
greater than the length of the cylinder. The mould, bound together
with split rotan, is placed centrally and vertically over the projecting
rotan, thus forming a box closed below with clay, open at the top,
and having a rotan in the centre. Into tliis the molten metal is
poured. Wlien cool the rotan is withdrawn, the mould open, and
the cylinder is complete. A good mould will make three or four
castings, but, as a rule, the first destroys it. The measurements of
the cylinder are : — length, 3| inches ; width, | inch ; bore, | inch.
This is the average size ; larger ones do not work well, smaller ones
are of no use/

BRmsH North Borneo, — ^The only actual specimen wliich I have
from British North Borneo is one (fig. 37) which was sent in 1890
by Mr. L. P. Beaufort, who collected it on the west coast, to Sir
R Biddulph Martin, who has very kindly given it to me. It is quite
a remarkable and specialized form, unlike any other which I have
seen from any part of the East, As in the last-mentioned examples
from Sarawak, the cylinder is of lead, or possibly lead and tin, cast
in a bivalve bamboo mould, and decorated at the lower end with
faintly raised, foliated designs, and at the upper end with punched or
incised zigzags. The great peculiarity of this example lies in the
form of the lower end of the cylinder. The base, instead of being
flat or rounded, is of unsymmetrical form and concave, and just above
this is a broad, rounded notch on one side. From this notch a per*
forafed duct cammMmcates with the bottom of the lx)m of the cylinder ^ very
much after the fashion of the touch-hole and fire-duct of an early
muzzle-loading cannon. The presence of this duct is a most pecuhar
feature, and its raison d^etre is not readily accounted for. It certainly
recalls to one's mind those early European and English forms, in
which the air is violently driven through holes, to which I have
already referred, and it has occurred to both Mr, Miller Chiisty and

THE FIRE-PISTON

myself that, possibly, the tinder was held in the outside notch against
the small orifice, through which the air was violently driven in a
compressed state by the piston, the friction due to passing through
the small duct being largely responsible for the production of heat.
At the same time I am dLsinclined to think that this was the case.
The duct is, to my mind, far too large for the purpose, and it does not
appear to have been enlarged at all since it was first made ; through
such a duct the air would escape so easily and quickly when forced
through by the piston that there would be very little compression
or friction, and, consequently, very little rise of temperature. The
tinder, moreover, would almost certainly be blown away. It seems
to me more likely that the tinder was placed, as usual, on the
end of the piston (which is, indeed, hollowed out, cup^wise, in the
usual manner, evidently with this intention), and that when the piston
was driven forcibly downwards, the small orifice of the duct was
tightly closed by a finger which would lie comfortably in the rounded
notch. This would allow the air to be compressed, as the cylinder
would thus be, temporarily, a dosed one. At the end of the piston
stroke, when the tinder was ignited, the finger would be raised, thus
opening the duct, and, in addition to the piston being more readily
and quickly withdrawn, through no vacuum being formed, the air
from the outside, which would rush in through the open duct owing
to the auction of the piston, would actually blow up the tinder into a
higher state of incandescence, rendering it unnecessary to blow upon
it after removal from the cylinder. I offer this theory as a possible
solution of the mystery of this peculiar type, though as yet I have
not been able to conduct experiments in order to see if such a pro*
oees would act efficiently. The piston of this specimen is of wood,
and presents the peculiarity of the cupped end having been capped
with lead. This lead capping is damaged, and it is not easy to see
whether it was mtended to take the place of a packing or whether
it was supplementaiy to the more usual packing of thread. No
trace of thread packing is to be seen, though a sunken groove
near the end of the pistons seems to be designed for holding some
kind of packing wound round at this point Mr. Beaufort told
Sir R. B, Martin that fire-pistons were becoming very difficult
to obtain in British North Borneo, where they are confined to
the west coast. He also added that ' the better ones are made
of wood V

' Joum. Anthrop. Inst., xx, 1891, p, 33 L
D 2

THE FIRE-PISTON

The only example made of wood from Borneo is one figured by
C- M. Pleyte/ and, although this is not so stated, it seems likely that
this may have come fi'om British North Borneo, It is (fig. 38) quite
plain, and differs in external detail from examples fi:-om Sarawak.

In regard to the general question of the presence of the fire-
piston in Borneo, it appears to be confined to an area extending from
the westerly portions of Sarawak to the western coast of British North
Borneo, though there is a wide hiatus in the distribution between
these two regions. It is only found on or comparatively near the
coast, where there is a strong admixture with the Malay element, and
where Malayan cultui-e is veiy evident Both Mn C. Hose and Mr,
R Shelford are strongly of opinion that this instrument has been
introduced by the Malays, from whom the Sea Dayaks have borrowed
it in comparatively recent times. Mr. Shelford wrote to me in
answer to my inquiries that *the Malays and Sea Dayaks of the
Saribas Kiver were at one time associated a good deal in piracy, &c.j
and there was a good deal of intermarrying ; at the present day the
'* Orang Saribas" have more of the Malay in them than any other
tribe of Sea Dayaks, and, as far as I can make out, they are the only
tribe who know the use of the cheiop {L e, fire-piston) \ The latter
remark leaves out of consideration the occurrence of the implement
in British North Borneo; but there, too, Malayan culture is not
lacking on the coast, and it is likely that the forms found there,
wliich differ from the Sea Dayak forms of Sarawak, are traceable to
the same Malayan origin, the difference in type being due either to
variation witliin the district or to different types of the instrument
having been introduced by the Malays. The use of lead as a material
is peculiar to Borneo, and it is possible that this may be a character
developed in the island itself, unless the Malays may have them-
selves used this metal and introduced its use with the instrument
itself. Of this there appears so far to be no record. Tliere is no
Siamese influence in Borneo, so that the direct influence of Siamese
culture from the Malay Peninsula is quite improbable.

jAVA-^Hre-pistonSj though now scarce in Java, range over
a wide area of the island. They are apparently always made through-
out of buffalo-horn ; at least, all the specimens I have seen or know
of are of this material.

^ Oldilms, UXf pt iy, p. B (of reprint), fig« 7.

THE FIKE^PISTON

A good, well-made specimen in my possession (fig. 39)j of black
horn carefully polished, has a cigar-shaped cylinder, with two bands
of ornamental engraving. The piston terminates in a large rounded
head, which is fixed to it with a horn rivet. This knob or piston-
head is hollowed out, and serves as a receptacle for tinder, which
consists of a brown palm-scurf. The specimen was obtained in
Buitenzorg in the west of Java. This shape appears to be a charac-
teristic one. Mr. C. M. Pleyte, of Leiden, had several examples of
this form from Bogor^ one of which is now in the Edinbui^h Museum;
these are almost identical with my specimen. In the museum at
Rotterdam there is a hom fire-piston from Java, but I do not know
if its shape is the same as the above. In the Cambridge Museum
may be seen a specimen from Kadiri (Kediri), in which the cylinder
is shorter and terminates in a smaU projecting knob. It is ringed
all over with transverse, incised lines (fig. 40). A diflferent type
again is figured by C. M, Pleyte/ in which the horn cylinder tapers
from below upwards, the base being broad and cut oiF square. The
knob on the piston is hoUowed for containing tinder, and is furnished
with a lid which fits over a flange (fig. 41). In the same article
Pleyte refers ^ to a Sundanese fire-piston (West Java) called tjeletok
Tlie form of this is, unfortunately, not described. He says that tjeUtok
m from the root word tjetok = Malay tjaiok ; nwntjatok = struck down
quickly or with force. The word is the same as tjatoew given as the
Malayan name of the instrument in Sumatra.

Flores. — Prom this island there is a fire-pkton in the Vienna
Museum (fig. 42). It is made of horn, and is peculiar in having a
rounded receptacle for tinder at the lower end of the cylinder, instead
of in the knob of the piston.

John Cameron says, as quoted above, that prior to 1865 he saw
the fire-piston in use in some of the islands to the eastward of Java^
so that we may assume tliat other islands in the neighbourhood of
Flores possessed the instrument at that time. He, unfortunately,
does not specify the localities.

Phiuppine Islands. — The fire-piston as it occurs in the Philip-
pines appears to be restricted mainly to the wild non-Negrito tribes

* Globus^ lix, pt. iv, p. 3 (of reprint).

* Quoting the catalogue of the Bataviaaache Genootsehap van Kunsten en
Wetenschappen^ p. 66, no. 1120.

THE FIEE^PISTON

of north central Luzon, where it is used by natives of the so-called
* Indonesian' group. It is also recorded from Mindanao, however.
H, Savage Landor says,* * This instrument called Bant in, generally
made of carabao horn, is found among various tribes of North Luzon,
and also in South LuzoUj among the curly-headed Aetas of the Gulf
of Ragay, , • / He does not specify the particular tribes in the north,
and it is unfortunate that he does not say if his information regarding
the Aetas is firstrhand or not I have found no other references to
fire-pistons among tribes of Negrito stock, and fiu-ther information is
required on tliis point A. E. Jenks remarks ^ that * the fire-syringe,
common west of Bontoc Province among the Tinguian, is not known
in tlie Bontoc culture area'. Others extend the distribution into the
Bontoc area, and beyond it into the more central portions of the
interior of North Luzon, Dr, Schadenberg mentions ^ then- use by
the Bontoc people, and describes the cylinder as of cambao (buffalo)
hom*tip, c. 9 em. long, with a bore of about 1 cm. The fire-piston,
together with a box for grease and tinder of charred cotton, is carried
in a pouch woven from bejuco. He adds that the natives value them
very greatly and require a high equivalent in exchange.

In the Dresden Museum there are two specimens. Of these,
one, from the Igorrotes of Bontoc (fig. 43), has a cylinder of wood
tapering from below upwards ; the other (fig, 44), from the Igorrotes
of Tiagan, is very similar but is made of horn. Each has a separate
tinder*h older of bamboo.* Another Igorrote example (fig. 45), col*
lected by Dr. Alexander Schadenberg, is in the Vieima Museum.
The cylinder is of carahm horn and the piston of wood ; the tinder
of cotton is contained in a bamboo holder. The collector refers to
the use of the instrument among the Igorrotes of Tiagan^ Lepanto,
and Bontoc. P. H. Sawyer* gives the Igorrote name of the fiie-
piston as pamiguin. Sulpakan is mentioned as the native name of
a specimen from Luzon in the Berhn Museum, A Tinguian speci-

' Gems of the East, 1904, ii, p, 334.

' The Bontok Igorrote MamJa, 1^5, p* 134. (Department of the Interior^
Ethool. Surrey Publications, vol. i.)

' Vcrhandl d. Berliner GeseU, / Anthrqp,, 1886, p. [551], in Zeitf, Mfmol.,
vol. xviii.

* Publkaiimien aus dem Kgh Ethnog. Museum zu DresdeUy by A. B, Meyer
and A. Schadenberg, viii, Bie Fhilippinen^ 1. Kord Luson, 1890, p, 21, and pi. xriij
figt. ISand 19.

* The InJiahitants of the Phaippincs, 1900, p. 266.

THE FIRE-PISTON

men is in the latter museimi. In tlie Ethnological Museum at
Rome there is a fire-piston from the Calinga tribe in the province of
Nueva Vizcaya, collected by Jose M** de Mourin, 1893 (fig. 46), The
horn cylinder is longitudinally faceted and transversely ringed at
either end The piston is of wood. D. C. Worcester mentions *
examples made of buffalo-horn from the wild tribes of North Luzon.
He adds : * To perform this operation successfully requii-es long
practice. I have yet to see a white man who professes to be able to
do it. • , - How the savages first came to think of getting fire in such
a way is, to me, a mystery.* I may assure him that the pmcess of
procuring fire by this means is quite easy, provided that the bore of
tlie cylinder is true and the piston cai-efidiy packed. In Mr. Edward
Bid well s collection there is an example (fig. 47) from Luzon with
horn cylinder and wooden piston, made very plainly, Mr. Landor *
says that in the more elaborate fire-pistons from Luzon * a receptacle
for the tinder balls is to be found and a metal spoon attached '.

Lastly, there is a reference to the fire-piston in Mindanao, the
southern island of the Philippine group. F, H. Sawyer mentions ^
it as being used by the Mouteses or Buquidnones in that island.

Origin and Bi^tsal

Having given as far as my present information admits a
description of the geographical distribution and varieties of the fire-
piston, let me now turn to the more difficult though perhaps more
interesting side of my subject. The question arises, What do we
learn as to the history of this instrument from its distribution ?

The two regions m which it occurs are very widely separated,
both geographically and culturally. On the one hand, we have
Western Europe and England as a home of the fire-piston in an
environment of the highest culture ; on the other hand, we find it
occurring over a very wide but very connected area in the East,
amongst peoples relatively low in the scale of civilization. The
primary question requiring solution is whether the fire-piston has
been tratistnitted from the one geographical area to the other, or
whether it was independently arrived at in the two regions. We
know that the principle of the method of producing heat by com-
pression of air was discovered in England and France by scientific

» 'Flie Thilmnne Islands, 1898, p, 297.
* Op, cit, p. 345.

Op. ciL

THE FIKE-PISTON

experiment, and that this principle was to some extent adapted to
domestic use there, by the invention of the fire-piston, so that it is
at least clear that the European form was not derived from the East.
Was, then, the Eastern instrument a derivative from the Western ?
This question is not easily answered. On the one hand, the
difficulty of explaining how native peoples, in a comparatively low
condition of culture, could possibly have arrived independently at
the knowledge requisite for the invention of this method of fire
production is so great as almost to compel the belief that the
instrument fntist have been introduced from elsewhere by some
more highly cultured race. It must be remembered that it is only
100 years ago last February that the first English patent was taken
out by Lorentz for a fire-piston, and that the scientific knowledge of
this method of obtaining a spark dates only from a very few years
earlier. This, among a people in the highest state of civilization
and of scientific advancement. It seems almost incredible that so
delicate and far from obvious a method can have been discovered,
whether by accident or by gradual development, by any of the
Eastern peoples amongst whom it has been found in use. At the
same time, it must be admitte<l that this is the o«% serious difficulty
which lies in the way of admitting the possibiUty of an independent
origin in the two main regions of distribution. Ther© is no inherent
impossibility in such a double origin ; cases of independent in-
vention of similar appliances in widely separated regions having
frequently arisen. There is no record of introduction by
Europeans.

There are, furthermore, considerable difficulties in accounting
for the dispersal of the fire-piston in the East^ under the theory of
its original introduction from Europe. From the earlier references
we learn that prior to 1865 the fire-piston was already well known
in the East over a very extensive geographical area^ embracing
Burma, the Malay Peninsula, Borneo, and the * islands to the east-
ward of Java \ This is a wide range of distribution, and it would
seem probable that considerable time would be required to account
for this extensive dispei*sal, even if the instrument had been intro-
duced by travellers from the West, If we choose to conjure up
a picture of enterprising European voyagers in the earlier half of
last century depositing supplies of fire-pistons in various islands
of the Malay Archipelago and on the mainland of S*E. Asia, we
must also allow for the time which must have elapsed before due

THE FIRE-PISTON

appreciation of the value and potentialities of the new machine
would have been developed in peoples to whom its principle was
hitherto absolutely unknown. We must also allow for a still longer
perio<l during which the difficulties of making imitations of the
European instiiiment by native methods were gradually overcome j
for we must bear in mind that, simple and few as are the essential
elements which together form the fire-piston, it is only when they
are in perfect adjustment that the instrument will work effectively
and produce the desired result. To this extent the fire-piston is
[essentially a delicate instrument ; an imperfect bore, faulty packing
[of the piston, or inferior tinder, will at once render the appliance
practically useless. Native made and effective fire-pistons were
certainly widely distributed in the East before 1865. European
travellers who observed them expressed great astonishment at this
peculiar method of fire producing, which was evidently quite new to
them. They were educated and experienced men, and we may

ither from their marvelling at the method, that they were un-
^acquainted with it at home, where the domestic use of the fire-piston
must have long since died out, Bastian, who records in 1866 that
he had seen the fire-piston in Burma, was born in 1826, and was
therefore about forty yeai-s old at the time, and although Iiis memory
would have gone back so far into the early half of the last century,
he was evidently unfamiliar with the instrument in Europe, It is
unlikely, therefore, that the instrument was of at all recent intro-
duction from Europe at that time. Another important point to be
remembered is the fact that no fire-pistons of European make have,
apparently, been found in the Eastern area of dispersal.

From the passage in the Mechanics' Magazine quoted above, we
may gather that in 1842 the fire-piston was but little known in

Cngland, though it is said to have been familiar on the Continent,
tt appears on the whole unlikely that this instrument can have been
taken out as a trade article to the East by English travellers later
than say 1830, since its practical use, never very prevalent in
England, seems to have been quite on the wane by that time. Nor
is it likely that it would have been traded abroad much earlier than,
say, 1816, since its first introduction to domestic use in England
was no earlier than 1807. This would allow a probable maximum
period of fifteen years during which English traders and travellers
could introduce it to various parts of the East. The predominant
European influences in those regions which are comprised within

THE FIRE-PISTON

the area of dispersal of the fii-e-piston in the East, have been the
English and the Dutch. Of the use made of the instrument by the
Dutch, I have no record, but at least it would appear that they were
not very vigorous in pushing tliis article in the Malay Archipelago,
since such large possessions as Dutch Borneo, Celebes, and the
Moluccas, do not appear to have received the instrument. As to
the French, who appear to have entertained a kindly feeling towards
the fire-piston and to have made fairly considerable use of it, they
need hardly be considered as possible introducers, since the regions
of geographical distribution of the fire*piston in the East are mainly
outside the sphere of their direct influence.

It is certainly difficult to account for the wide Eastern distri-
bution of the fire-piston and the development of local native varieties
by the theory of introduction from Europe, which allows so short
a time in which to develop the conditions which already obtained
prior to 1866. This is especially the ci^e when we remember that
such primitive and widely separated peoples as the Mois of Indo-
China and the Indonesian peoples of Luzon in the Philippines are
well acquainted with the manufacture and use of the instrument.
These peoples have until recently been very little known to
Europeans.

It may be suggested that Europeans may have introduced the
fire-piston into some one or two districts only^ and that the further
dispersal was eflfected by transmission elsewhere through native
agency. This would, however, have required a longer time than is
available, as dispersal by this means is necessarily slow.

It has frequently been suggested that the Chinese must have
originated and organized the dispersal of the fire-piston in the East.
It is a common practice to credit the Cliinese with the invention of
many strange things, but there is, unfortunately, no evidence what-
ever that they even knew of the fire-piston, except perhaps on the
Burmese and Siamese frontiers. At least, as far as I know, there
are no records or specimens which give evidence of such knowledge
on their part.

The geographical distribution of the fire-piston in the Siamese
Malay States and the Malayan regions of the Peninsula has caused
some of the distinguished local experts to believe that the instrument
is rather Siamese than Malayan in origin, as far as that region is
concerned. This theory would perhaps account for its north-
easterly and north-westerly dispersal amongst the Mois, the Slians,

THE FIKE-PISTON

and the Kachins. It is possible that the Malays may have borrowed
it from the Siimiese. Be this as it may, the Malays have certainly
acted, perhaps not as the sole, but at any rate as the main, dispersers
of the fire-piston over the islands of the East Indian Archipelago^ from
Sumatra to the Philippines. Wherever in this region the fire-piston
is ibund — even though it be in the hands of and manufactured by
more primitive peoples— the influence of Malayan culture is also
observable, and the instrument is not found in districts which are
remote from Malayan contact. It is even possible that the Malays
are the actual originators and that the Siamese may have borrowed
the idea from them* Or the evidence of its frequent use amongst
the widely separated ' Indonesian ' or Proto-Malay tribes of Luzon
and the Mois of Indo-China, who are by some ethnologists classed
as belonging to the * Indonesian ' stock, togetlier with the fact that
the neighbouring more highly cultured peoples are without it, may
be taken as pointing to a Proto-Malayan origin, which would assign
the invention of the fire-piston to a race still lower in culture than
the Malays proper. This theory would involve a very considerable
antiquity for the Eastern fire-piston and the probabilities are i^erhaps
hardly in favour of it All that can be said with any certainty is
that, whether the fire-piston was introduced to the Malays by
Europeans or by some other Eastern people in a condition of culture
more or less on a par with their own, we must, I think, give to
the Malays due credit for having materially assisted in extending the
geographical range of the instrument and of having introduced it
into several of the islands of the Eastern Archipelago where it has
taken root, and where local varieties have in the coui^se of time
arisen and themselves again become modified in matters of detail.

With the single exception of the peculiar type from Britisli
North Borneo (fig. 37) all the Eastern forms are essentially the same
in general structure, the less important details being those which
alone are capable of modification and variation. Such are the
materials used in the manufacture of the cylinder and piston which
may be of bamboo, wood, horn, ivory, bone, brass, or lead (lead
and tin usually} ; the external fonn ; such accessories as the tinder
receptacle which may be separate from the instrument, and consist
of bamboo, nut-shells, beans, palm-spathe, or of woven materials.
Prickers for adjusting the tinder, grease-boxes and spatulae for
applying the grease to the piston-packing, are other accessories
which may be present or absent, but whose occurrence in identical

THE FIRE-PISTON

shape in widely separated regions adds to the evidence which goes
to prove that the whole series of Eastern types belongs to one
morphological group.

Assuming, for purposes of argument, that the Oriental fire-piston
was invented independently by the relatively primitive peoples
amongst whom it appears to have been in use during a long period,
we may consider the question as to the manner in which these
people might conceivably have hit upon this highly specialized
method of producing fire. It must be admitted that the great diffi-
culty in arriving at a satisfactory conclusion upon this point is the
principal factor which militates against the acceptance of the theory
of the native origin of the fire-piston. There can be Uttle doubt
that, if the invention was made by an Eastern people, the principle
must have been arrived at by some happy accident, the effect having
been produced during the process of some action or work uncon-
nected with fire-making. It is inconceivable that such a physical
phenomenon could have been thought out and elaborated scientifi-
cally by primitive peoples, and we may remember that in Europe
the first appreciation of this phenomenon of heat-production by air-
compression was due not to research but to observation of an
unexpected effect. There are three absolute essentials necessary for
production of heat in this manner : (1) a cylinder with accurate bore,
closed at one end ; (2) a piston accurately fitting the cylinder ; (3)
tinder which is very quickly inflammable- Therefore, in our search
for prototypes we are necessarily restricted to objects in which these
elements may conceivably be associated.

A form of bellows used in blowing up the fire, which is very
prevalent in Burma and many parts of the mainland and the Eastern
Archipelago, and which belongs largely to Malayan culture, is con-
structed upon the principle of a piston ; there is a cylinder and a
packed piston, whose thrust drives the air out in a forcible manner.
In this, however, a duct opens from the lower end, and since, therefore,
the cylinder is not a closed one, there can be but little comprESsion
of the air ; certainly not sufficient to cause a marked rise in the
temperature; So that even if by accident some tinder-like material
adhered to the piston, it could not be ignited. In breaking through
the nodes of a bamboo, in order to render the bore continuous and of
greater holding capacity, a rod may be thrust violently down the
cylinder which at fii^st is, of course, closed. Certain simple and
primitive-looking fire-pistons amongst the Kacliins are indeed

THE FIRE-PISTON

made of natural bamboo cyKnders. It is unlikely, however,
that the rod would fit so tightly as to act like a jmcked piston, and
hence there would be next to no air-compression. Appliances of
the nature of toy pop-guns and water syringes are not unknown in
the East, but although these exhibit some structural resemblance to
the fire-piston, there seems little likelihood of their having suggested
the latter. The process of boring and gauging blow-guns when
these are made of solid wood might, conceivably, have led to some
unintentional compression of the air within the bore, which might
have caused the ignition of some responsive material adhering to the
boring- or gauging-rod* While even this is improbable, it is interest-
ing to recall that the distribution of the Oriental blow-gun embraces
many of the regions where the fire-piston is found. I have
frequently had it suggested to me, that it is obvious that the fire-
piston must have been derived from the pestle-and-mortar so
commonly used throughout the Indo-Chinese and Malayan area for
crushing the betel-nut or chavka leaves. In favour of this, it may
with truth be urged, that there is often a very strong resemblance
between the two appliances ; indeed some of the small pestle-and-
mortar apparatus in the British Museum bear so striking a resem-
blance to some of the Bornean fire-pistons, e,g. the type shown in
fig, 36, that it is necessary to look carefully at the specimens, in
order to see to which group they belong. On the other hand, it
is evident that the suggestion that the pestle-and-mortar is the
prototype of the Eastern fire-piston is based solely upon this super-
ficial similarity, which is evidently appreciated by the Malays, since
they apply the word gohek to both instruments. We have only to
remember that for all practical purposes, characteristics which are
essential to the efficiency of the one instrument are absolutely detri-
mental to that of the other. In the case of the fefe?-mortar, it is
imperative that the pestle should work loosely in the mortar, and
it is equally essential that in the fire-apparatus the piston should very
accurately fit the bore. A slight departure from this rule in either
case renders the instrument useless for its purpose, and it is, conse-
quently, most improbable that either could have accidentally per-
formed the function of the other and so have suggested it.

One other appliance seems to have a claim to consideration. In
the process of cleaning the barrels of the small muzzle-loading
cannons, such as are frequently seen in the East Indies, it is con-
ceivable that in driving an accurately fitting cleaning-rod up the bore

THE FIRE-PISTON

with some force a considerable compression of the air inside might
result, and that a piece of readily combustible matter miglit have
been ignited thereby. The touch-hole, being very small, might not
have caused a too great diminution of the air-pressure, since the air
could only escape relatively slowly through this orifice ; or on some
occasions the touch-hole may have been temporarily blocked, in
which case the compression would have been greater and more
effective.

In some respects this appears to be the least unlikely of the
possible suggestions as to the prototype of the fire*piston, and colour
is lent to the idea by the form of the North Borneo fire-piston
(fig, 37), in which the cylinder has the appearance of a miniature
cannon actually Jitted with a * toucMwle \

At the best, however^ I am not at present able to offer any very
convincing suggestions as to how the fire-pLston may possibly have
been discovered in its Eastern home, and it seems all too likely that
the question of its nmnogenesis or polygenesis may never be completely
determined. The problem remains an exceedingly interesting one,
both from technological and ethnological standpoints, and, in con-
cluding this attempt to bring together the material available for
comparative study, I may express the hope that further information
may be forthcoming, both as regards the earliest records of the fire-
piston in the East, and as regards the geographical distribution and
varieties of this pecuHar method of producing fire.

I wish to thank heartily those who have so kindly assisted me to
procure specimens or information. More especially am I indebted
to Messrs, Skeat, Annandale, Shelford, Leveson, Miller Christy,
Joyce, and Bidwell, whose assistance has been of much value to me.

LIST OF ILLUSTRATIONS

Fig. L Fire-Byringe, from Patent Specification of Eichard Lorentz, 1807,
No. 3007 ; prmted 1856.

Fig. 2. Fire-piston, from E. J. Mitchell, June 19, 1882, in The Medianics"
Magazine^ xvii, 1832, p. 828.

Fig. 3, Fire-piston, Franca From Tiie Penn^ Magazim, July 26, 1834, p. 268.

Fig. 4, Fire-piston, England ; of rolled brasa ; length of cylinder, 14 cm. For
domestic nae or for scientific experiment E. Bidwell collection.

Fio* 5. Ditto, England ; cylinder of rolled brass, 10-2 cm. long ; piston of steely
9*§ CM., with brass mounts and leather packing. E. Bidwell collection.

THE FIRE-PISTON

Pig. 6. Ditto, England ; cylinder of cast brass, 8 1 cm. long ; piston of ateel,
8-6 cm*, with brass mounts ; the packing is of brass, E. Bid well collection.

Fig* 7. Fire-piston, modern French ; cylinder of white metal, 7-6 cm., with
ebony knob ; at side, a tube for cord-tinder fitted with ball-and-chain extinguisher ;
piston of ebony, 7*8 cm. Parchased in Paris* Given by Mr. Christy Miller to
author, 1902,

FiQ. 8. Fire-piston, Kachin, Northern Shan States, Lat. 24° T N., Lon. 98°
15' E. J cylinder of bamboo, 8*1 cm, ; piston of wood, 9*2 cm. Given by Mr. H.
Leveson to author in 1898,

Fig. 9. Ditto, same data ; cylinder of bamboo, 8 cm. ; piston of wood, 13-1 cm.

Fig. 10. Ditto, same data ; cylinder of bamboo, 8*5 cm. ; piston of wood,
11*5 cm., carved head,

Fio. IL Fire-piston, made hy Wa villagers on the Chinese frontier of East
Burma, 22° 20' N,, 99° 10' E. ; cylinder of stout bamboo, 7^2 cm* ; piston of hard
wood, 10*9 cm. Given by Mr. H. Leveson to author in 1900,

Fio. 12, Fire-piston, Ruby Mines, Mogok, Burma ; cylinder of lathe-turned
bamboo, 8*9 cm. ; piston of wood, 16 cm. Obtained by Mr, Frank Atlay and given
by him to author in 1907.

Fio, 13. Fire-piston, Kaijb, Cauri Kachins, East of Bh&mo, Upper Burma ;
cylinder of light*co loured horn, c. 76 cm. ; piston of wood, c, 9-5 cm. Collected
by Leonardo Fea, 1885 ; Ethnological Museum, Kome [40232],

Fig. 14, Ditto, same data ; cylinder of black horn, c. 8»5 cm, ; piston of horn
riveted to wooden knob, Fea collection ; Ethnological Museum, Rome [40233],

Fio, 16, Ditto, Kachins of mountains East of Bhamo ; cylinder of black horn,
e. 8*2 cm. ; piston of horn riveted to wooden knob, Fea collection ; Ethnological
Museum, Rome [40285],

Fig. 16. Ditto, Kachins and Shans in mountains East of Bhamo ; cylinder of
black horn, c. 9 cm. ; piston of horn riveted to wooden knob, Fea collection ;
Ethnological Museum, Rome [404 72 J.

Fig, 17, Ditto, Kacbin (Simpfo), Bhamo district ; cylinder of black horn,
7*9 cm. ; piston of horn riveted with horn to wooden knob, 12 cm, Fea collection,
1885 ; given to the author by Prof, E. H. Giglioli, 1903.

Fig. 18. Fire-piaton, obtained from a Kachin on the Chinese border of the
Northern Shan States, 24° 7' N., 98^ 15' E. Collected by Mr. H. E, Leveson, 1898,
and given to the author.

Fig. 19. Fire-piston, Kachin, Upper Burma ; carved cylinder of black horn,
8*6 cm, ; piston of hard wood with knob wrapped in silk, 13*3 cm. Collected by
Captain R. C. Temple and given by him to the Pitt-Rivers Museum, 1890.

Fio. 20. Fire-piston, Kachin, Upper Burma ; lathe-turned cylinder of black
horn with silver mounts, 6*3 cm. ; turned piston of horn with brass-ringed knob ;
attached to it are a bag of velvet and silk containing vegetable-floss tinder, a grease-
box of nut-shell, and an ivory spatula for grease. Collected by Major K. C. Temple
and given to the Pitt- Rivers Museum, 1894.

Fig, 21. Fire-piston, Mandalay, Burma ; lathe-turned cylinder of black horn,
6*4 cm. ; piston of horn riveted to turned horn knob ; attached to it are a cloth bag
with vegetable-floss tinder, and a spherical, lathe-turned wooden box for greaeo.
Given by Mr. H. O. Mordaunt to the author, 1899.

THE FIEE-PISTON

Fig. 22. Fire-piston, Ruby Mines, Mogok^ Burma. Collected by Mr. Frank
AUay. From a sketch by Mr. B* Gunn.

Fia. 23. Fire-pision, mi-put, obtained from a ptm^ at a monastery, Sontbem
Shan States ; cylinder of hard wood, 11-9 cm. ; piston of black horn, 13- 1 cm.
Collected by Mr. H. Ijeveaon and given to the author, 1890,

FiQ. 24. Fire-piston, mi-put, Southern Shan States j cylinder of black horn
gracefully flutedj 8-5 cm. ; piston of black horn with knob inlaid with metal pins,
15 5 cm. ; furnished with a tinder-pouch of palm-apathe, and a turned wooden
grease*box (fig. 24 a)* Collected by Mr. H. Leveeon and given to the author, 1891.

Fig. 25. Fire-piston, Khas or Kumuks, North of Luang Prabang, Siam ;
cylinder and piston of horn ; with bag of vegetable-floss tinder. Science and Art
Museum^ Edinburgh.

Fig. 26, Fire piston, gopi api (Malay) or lek-phoi4ok (Siamese), Ban Sai Kau,
Nawnchik, Patani, Siamese Malay States ; turned cylinder of black horn, 5-7 cm, ;
piston of horn, 6-5 cm. Collected by Mr. Nelson Annandale and given to the
Pitt-Rivers Museum, 1902.

Fig. 27. Ditto, same data ; cylinder of dark horn, lathe-turned, 8 cm* ; piston
of wood fitting into horn knob. Annandale collection j Pitt-Rivers Museum.

Fig. 28* Fire-pieton, gohek api, obtained from Malays in Jalor, Patani, Siamese
Malay States ; cylinder of light horn, 6-8 cm. \ piston of wood. CoUected by
Mr. W, W. Skeat ; Cambridge Museum.

Fig. 29. Ditto, same data ; cylinder of turned bone, 96 cm. ; piston of wood.

Fig. 80, Fire-piston, from the Samsam village of Ban Phra Muang, Trang,
Siamese Malay States; cylinder of light horn, lathe*turned, 7 cm. ; piston of turned
black horn with knob hollowed out for holding tinder, 7*8 cm* ; Annandale collec-
tion, 1901 J Pitt' Rivers Museum.

Fig. 31. Fire- piston, obtained by Mr. Henry Louis on the Ayer Katiah, a small
tributary to the Teluban River (this presumably is the Telubin River in Patani),
Malay Peninsula ; cylinder of wood covered with cane-work rings, 6*5 cm. ; piston
of hard wood ; tinder-box (fig. 31 a) made from an tnluda bean. Given by Mr* F, W,
Rudler to the British Museum, 1901.

Fig. 82. Fire-piston, Malays of Soepajang, Menangkabau, Sumatra ; made of
buffalo horn, cylinder a U cm. long j copied from Veth, ' Midden Sumatra/ 1877-9,
pt ii, pL Ixxxiii, figs, 12 and 13.

Fig. 83. Pire-plston, Sumatra ; copied from E. T. Pritchett, * Smokiana,' p. 97.

Fig. 34. Fire-piston, Fort van der Capelle, North Padang, Sumatra j carved
cylinder of dark horn, 8 2 cm. \ piston of horn, carved, and with knob hollowed
out for tinder and fitted with ltd which, with a half-turn, can be secured by a pro-
jection which passes through a notch (fig. 34 a). Collected by Mr. Carl Bock ;
British Museum.

Pig. 85^ Fire-piston, gochok api (Malay), pafdang hesi api (Sea Dayak), Sea
Dayak, Simanggang, West Sarawak ; cylinder of brass lined with lead, 9-1 era. ;
lathe-turned piston of vpood ; cunarium nut with vegetable tinder and brass pricker
attached. CoUected by Mr. D. I. S. Bailey, and given by him to the Pitt-Rivers
Museum, 1904.

Fig. 86. Fire-piston, same data ; cylinder of lead (or lead and tin) cast in
bamboo mould, 8-1 cm. ; carved piston of hard wood, 11-6 cm. Bailey collection,
Pitt-Rivers Museum, 1904.

THE FIRE-PISTON

Pig. 37. Fire-piston^ west coast of British North Borneo ; of very unusuaJ con-
struction ; cylinder of lead (or lead and tin), cast in bamboo mould, with lateral
notch on one side at lower end^ from which a duct leads to the bottom of the bore
in the cylinder ; length of cylinder, 10-3 cm. ; piston of wood, 13 S cm., capped with
lead at the lower extremity. Collected by Mr, P. Beaufort, 189Q ; given by Sir
R Biddulph Martin, Bart., to the author, 1907.

Fig, S8, Fire-piaton, Borneo ; of wood. Copied from C. M. Pleyte, * Indone-
aisohes Feuei^zeug," Globus, lix, No. 4.

Fig. 89. Fire-piston, Buitenzorg, West Java ; of black horn j cylinder, 10-6 cm.,
engraved ; piston riveted to knob, which is hollowed out for vegetable-floaa tinder.
Collected by Mr* C. M. Playte ; author's collection.

Fjo. iO, Fire-piston, Kediri, East-central Java ; of black horn ; cylinder, 7 em.
Cambridge Museum ; Kgure taken from facsimile belonging to Mr. £. BidwelL

Fig. 4L Fire-piston, Java ; of buffalo horn ; the knob of the piston hollowed
and fitted with lid, forming a tinder-box. Copied firom C. M. Pleyte, * Indonesisch«8
Feuerzeug ', Ghbus, lix, No. 4.

Fio. 42. Fire-piston, Flores Island, East Malayan Archipelago ; made of horn ;
cylinder fitted with tinder receptacle at lower end. Vienna Museum ; from a rough
sketch.

Fig, 48. Fire-piston, Igorrotes of Bontoc, North Luzon, Philippine Islands ; of
wood, engraved j piston of wood ; tinder-holder of bamboo, 10 cm. Collected by
Herr C. Semper ; Dresden Museum ; copied from A. B. Meyer, PubL a. d. KmigL
Ethn, Museum zu Dresden^ viii, pL 17, fig. 18.

Fig. 44. Fire-piston, Igorrotes of Tiagan, North Luzon ; of buffalo-horn ;
cylinder, 8-8 cm. ; piston, 12*7 cm. ; engraved bamboo tinder-holder, 8 em. Semper
collection, Dresden Museum ; copied from same source, fig. 19.

Fig. 45. Fire- piston^ Igorrotes of Tiagan, Lepanto and Bontoc, North Luzon ;
cylinder of buffalo-hom ; piston of wood ; a bamboo holder with cotton-tinder
belongs to this. Collected by Dr* Alexander Schadenberg, Dresden Museum (S081S) ;
copied from a sketch kindly made by Irene Kust of Vienna.

Fig. 46. Fire-piston, Calinga tribe, Nueva Viscaya, North Luzon ; faceted
cylinder of horn, €. 6-3 cm. ; piston of wood. Ethnological Museum, Home (49164) ;
from a rough sketch.

Fig. 47. Fiie-piston, Luzon, Philippine Islands j cylinder of black bora,
7-5 cm. ; piston of wood. E. Bidwell collection.

Plate I

iip|

■±'

(S^ s>

1!^.

'H'-'l,

Icrpp)

Fire-piston: 1-6, Europe; 7-12. 18, 19, Further India.

EXOGAMY AND THE MATING OF COUSINS

By a, E. CKAWLEY, M,A, RRAJ,

Author op *The Mystic Rose, A Study of Primitive Marriage',
AND * The Tree of Life, A Study of Religion ', etc.

The problem of the origin of Exogamy has been narrowed
down within the last few years. In the first place, it is now
indisputable that Exogamy is a phase of a tendency, constant in
all stages of culture, towards out-breeding— mi unfortunate term this,
as the crucial question is psychological, and pre-scientific ideas of
* instinct * or * Nature's promptings * must not be allowed re-entrance
under the aegis of a eugenic pliilosophy. In the second place, we
have fixed upon the practical starting-point of this tendency, in the
prohibit hn against the mating of brother and sister.

We must first note that such a prohibition could not have origin-
ated, in the first instance, on the ground of kinsliip. Are we then
to accept the view of Messrs* Atkinson and Lang, to the effect that
(if I may quote Mr. Thomas's account) * men originally lived in isolated
groups, ruled over by an old male, exactly as a herd of cattle is
ruled. Tliis involved the exclusion of the young males, for the
whole of the adult female population of the group formed the harem
of the old male. Then in process of time it became possible for the
young males to remain within the group, which was thus immensely
strengthened for offence or defence, but only on condition that they
went abroad for their wives. As time went on, this rule, imposed
by the old male, crystallized into an instinct, and, the rights of the
old male falling into decay at the same time, there arose the law
that no one might marry within the group in which he was born ' ?

In this description there is too much to be assumed, and it
bristles with fallacies, though it may show the true eugenic touch.
The return of the prodigal sons, engineered by their mothers, is
perhaps the weakest pohit of Mr, Atkinson's explanation of the
Primal Law, Again, why should the rights of the old male fall

E 2

52 EXOGAMY AND THE MATING OF COUSINS

into decajj at a time when, ex hypothesi, legal instincts were on the
make ? The moral revolution which the primal law needed to start
it is too improbable. Last, but not least, it is a psychological impossi-
bility for any ruh^ imposed by any male^ old or youngs or for that matter,
by any female, to crystallize into or become^ in any way whatever, an
instinct It is, by the way, a possibility that pre*human men lived
like cattle in * groups', but there are more probabilities in favour of
the view that man has always had a more or less monogamous fire-
circle, as the unit of his social oi^anization.

We need an explanation of the law against brother and sister-
unions, which is derived from a sounder psychology.

Mr. Havelock Ellis, the soundest psychologist of the day, and
himself the discoverer of many tendencies in the human mind, the
imderstanding of wMch will be an inestimable boon to the race,
remarks in his Psychology of Sfer, * The explanation of the abhorrence
to incest is really exceedingly simple. Any reader who has followed
the discussion of sexual selection in the present volume and is also
famiUar with the '* Analysis of the Sexual Impulse" set forth in the
previous volume of these Studio will quickly perceive that tJie normal
failure of the pairituf imtinct to manifest itself in the case of brothers and
sisters^ or of boys ami girls brought up together from infancy^ is a merely
negative phenomenon due to the inemtuhk absence tinder those circum-
stances of the conditions which evoke the pairing impulse. Courtship is
the process by which powerful sensory stimuli proceeding from a
person of the opposite sex gradually produce the physiological state
of tumescence, with its psychic concomitant of love and desire, more
or less necessary for matmg to be effected. But between those who
have been brought up together from childhood all the sensory stimuli
of vision, hearing, and touch have been dulled by use, trained to the
culm level of affection^ and deprived of their pote^iey to arouse the erethistie
excitement which produces sexual tumescence. Brothers and sisters in
relation to each other have at puberty already reached that state to
which old married couples by the exhaustion of youthful passion
and the slow usage of daily life gradually approximate. Passion
between brother and sister is, indeed, by no means so rare as is
sometimes supposed, and it may be very strong, but it is usually
aroused by the aid of those conditions which are normally requimd
for the appearance of passion, more especially by the unfamiharity
caused by a long separation. In i-eaUty, therefore, the usual absence
of sexual attraction between brothers and sisters requires no special

EXOGAMY AND THE MATING OF COUSINS

explanation; it is merely due to the normal absence under these
circumstances of the conditions that tend to produce sexual
tumescence and the play of those sensory allurements which lead
to sexual selection. It is a purely negative phenomenon, and it is
quite unnecessary, even if it were legitimate, to invoke any instinct
for its explanation. It is probable that the same tendency also
operates among animals to some extent, tending to produce a stronger
sexual attraction toward those of their species to whom they have
not become habituated.* (Evidence on this ix)int is quoted.) ^ In
animalSj and in man also when limng under primUive conditions^ sexual
attraction is not a constant phenomenon ; it is an occasional mnnifeMation
only called out by powerful stimulation. It is not its absence which
we need to explain ; it is its presence which needs explanation^ and such
an explanation we find in the analysis of the phenomena of court-
ship/ I have put in italics those parts of the passage which are most
emphatic and throw the matter into clearest relief. In an appendix
to vol. iii of his Studies Mr. Ellis has a valuable discussion on * the
Sexual impulse among savages \ in which he proves the occasional
nature and periodicity of this function. Anthropologists should not
ascribe to primitive man either unbridled lust or an infinite capacity
for satisfying it.

To the above solution of our problem, I subscribe, with due
allowance for other psychological factoi-s, which vary with the
cultm^ of the race, e. g. sexual taboo, the moral law, proprietary
jealousy.

But the exogamous tendency became a legal prohibition against,
in the first instance, the mating of brother and sister. This still re-
quires explanation. Why should a natural tendency require the force
of law to corroborate and justify it? I tliink there is a simple
explanation. In many departments of primitive life we find a naive
desire to, as it were, assist Nature, to affirm what is normal, and
later to confirm it by the categorical imperative of custom and law.
This tendency still flourishes in our civilized communities, and, as
the worship of the normal, is often a deadly foe to the abnormal and
eccentric, and too often paralyses originality. Laws, thus made and
with this object, have some justification^ and their existence may be
due, in some small measure, to the fact that abnormality increases
pari passu with culture. But it is a grave error to ascribe a pre-
valence of incest to the period preceding the law against it. A close
analogy may be seen in the primitive attitude towards property, and

54 EXOGAMY AND THE MATING OF COUSINS

in mattei's of respect and etiquette. Other examples I have given
in my books, The Tree of Life and Tlie 3Iystic Rose,

The prohibition next forbids the mating of 'tribal' brothers
and sisters. This famiUar point needs no explanation here. It
is due to 'tribal' solidarity and is engineered by identity of names.
Totemism has no importance whatever for exogamy except in so
far as it is a system of names denoting ' kinship \ The totem is,
from beginning to end» a surname satis pbntse. If it is worsljipped,
well, so is the surname of many a civilized noble house.

The second intention of this paper is concerned with the
mating of cousins. Here, as before, I will try to avoid circum-
ference interests and to aim at a centre.

A fact ignored by the discoverer of * exogamy ' is this, that,
while it forbids the union of brother and sister, some cousins and
so on, it is actually in-breeding of a close kind. All the facts tend
to show that primitive man relied for his wives on friendly arrange-
ments as a rule. Fi^om liis point of view, the ideal state of things
would be that every tribe should be dual, so that wives could be
obtained without friction or diflSculty. And this is precisely what
we find in many uncivilized peoples. The tribe is divided into
two * exogamous ' sections, or phratries ; marriage outside the tribe
is forbidden, and also within the phratry, but is commanded between
the two phratries. The mechanical operation of ' descent', paternal
or maternal, on the names, totemic or otherwise, makes the units
of a phratry 'brothers' and 'sisters'. This interesting arrangement
is no%v well known. ^ How is its origin to be explained? A deli-
berate bisection, as Mr. Lang has proved, is unthinkable.

No tribe was ever deliberately divided ; the bisection must have
grown out of some simpler bisection. Wliat was this ? Mi\ Lang
supposes two of his kine-like hoi*des, headed by the old males,
forming an alliance for matrimonial purposes.

This is correct, no doubt, but requires modification if the nature
of the primitive group was not as described by Mr. Lang.

* In New Britain they are called after the two ][H>wers of Good and Evil,
To Kabinana and To Kovuvuru, As descriptive terms for them we have Veve in
Melanesia, whicli means dmshn, and appears to have obtained the furtiier connota-
tion of * motherhood ' ; among the Kai^ns they have no names, but are described as
Pak'tte ('of descent from the father's side') and Mo-tee (^ of descent from the
mother's side '). In Fiji members of the two aides of the house in each faniily
are described as 'marriageable \ cone ubi tan ts. There is nothing totemic liere.

EXOGAMY AND THE MATING OF COUSINS

My view then of the two phratries is, that, as we find them^
thetf are two great families, in the second and wider sense of the
term, and that tliey sprang from two families in the narrower sense.
In other words they are the ^ sides of the house' in am great dual
famUy, These two original families intermarried^ this is the first
step, and continued to intermarry generation after generation.
Each was originally exogamous, and of course remains so because
the members of each bear the same name, and are therefore
* akin ', whether really so or theoretically matters not to the savage,
but as a feet they will be so related. The two phratries thus
come first

The phratry-names {pace Mr, Lang) are usually unintelligible,
and therefore probably older than the names of the smaller families
or totem-kins which compose the phratries. This is one indication
that the two phratries ai'e themselves also earlier.

Secondly, the toteniic sftmll families which muke up emh phratry
are younger braficlies of the original dual family which Jiave come in
through marriage of tmnien taken from other groups and giving their
names to their children. Such a family name would naturally
be nearer^ as it were, to those who bore it than the name of the
greater family of wliich they form a younger branch. Mr, Howitt
has observed that the totems are living names, part of the living
language, and invariably derived from natural objects found in
the tribal country ; the phratry-name is general, ' the totem-name is
in one sense individual, for it is certainly nearer to the individual
than the name of his moiety.'

The two phratries are thus developed by a natural growth,
and are not due to a deliberate bisection of an existing community.
Hiey are implicit in the first marriage, which is the nucleus of
the future community. The totem-kins are not subdivisions, but
younger branches of the old famiUes. Families of the one great
family cannot intermarry because they belong to that family,
and they marry into the other great family because it is *the
other side \

It may be asked, why two families? Well, two families are
needed in every nmrriage, the family of the husband and the family
of the wife. Why should they continue to intermarry? Why
not ? Wives are not easy to come by in early society except from
friends, and the pressure of external circumstances will set a
premium on such combination. But will not the two faniihes very

EXOGAMY AND THE MATING OF COUSINS

soon become too nearly * related'? They will become 'related',
but not too nearly, for the children who marry in every generation
will have different names, the one being that of the female side
of the dual family, and the other that of the male. The inter-
marriage of the two phratries is often obscured in the minds of
investigators by the prohibition to marry in the same phratry, but
in the native view it is just as important. Lastly, it is only
cousins who can marry, and as the earliest peoples have no term
for cousin, it is probable that this relationship was not originally
regarded as being more than sl friendly reUUion.

I suppose two friendly fire-circles, consisting each of father,
mother, and one or more children. It does not matter whether
the two are related or not. They will naturally exchange daughters
in marriage to their sons. This is the most usual method of
obtaining wives in Australia, and is I think the most primitive.
Thus we get two or more new fire-circles in the close neighbour-
hood of the old, the friendly relation will be emphasized by all
the circumstances of a nomadic life, and the two connected families
will keep together, I presume an exogamous tendency, already
explained, towards marrying outside of the tixe-circle, combined
^vith a preference to marry those of the same age. The next
generation will, so far as the balance of the sexes allows, marry
in the same way, this time cousins* They do not recognize any
real relationship in this as yet, as the earliest savages do not ; what
is always known, at least by modem savages, is the relationship
of parent and child, brother and sister. These people then may
be supposed to know who belongs to the two families. At any
rate as soon as names are applied there will be no difficulty in
distinguishing them. The system works both with male and female
descent, with either totemic or numerical, local or descriptive
names, nicknames or complimentary appellatives. With female
descent the two names will be dotted here and there ; with male
descent the holdera of one name will tend to be grouped together.
The latter state of things may end in local exogamy. There is
an important principle probably univei-sal in early times, that
a wife does not take the nmne of, or beeome kin to^ her husband. This
creates a perpetual potentiality of marriage between her side of the house
and her hmban<rs, and doubtless had much to do with delaying
the recognition of relationship between those cousins who have
different names. The two families will in the second generation

EXOGAMY AND THE MATING OF COUSINS

Bee themselves reprodmedj and also in the third and following, by
the two sets of intermarrjing cousins.

With regard to cousins and their mating let us note^ first, that
it has been proved that this union is by no means deleterious to
the offspring. Cousin-marriage is a well-known mark of dual
exogamy, but it occurs in a form which may seem strange^ if one
does not make a diagram. The peculiarity is that while the children
of two brothers may not marry, nor the children of two sisters,
the children of a brother and a sister may. This is an atUonmiic result
the fact that the name of the family is inlierited; it makes no
difference whether male or female descent is used.

Smith m.
Jones f*

Smith m.
JoEes f.

Jones m.
Smith t

Jones in.
Smith f.

II I I I I

Smith m. Smith f. Smith m. Smith f. Jones m. Jones f- Jones m* Jones f.
(cousins ; (cousins ; (cousins ;

may not may may not

many.) many,) marry,)

ITie children of the brother and sister Smith may marry because by
their names they belong to opposite phratries.

This peculiarity was first noted by Dr, Tylor, who called it cross-
cousin-marriage* All peoples who allow cross-cousin-marriage thereby
show that they recognize the two sides of the house^ and have the
;erm of the phratry system. Cousin*marriage generally is the most
'favourite connexion among early peoples. Mr, Fison says that
* in some parts of Ireland, at the present day, a girl will sometimes
reveal the state of her affections to the youth on whom she has set
her heart, by saying, "I wish I were your cousin." And this is
understood to be an offer of marriage.' It is what may be called
the * endogamous ' tendency, and the cousin-marriage termed cross is
a key to the phratry system. In the two-phratry system of the
Iroquois, each phratry is called a ' brotherhood ' ; the families of
phratry A are * brother'* families to each other, and ^cousins-families
to those of phratry B and vice versa, a case which, so far, proves
the whole business.

But how is this dual family, the nucleus of a possible tribe,

grow? It does not seem to have ever been pointed out that

cousin-marriage, and all such * endogamy ', tend to check the increase

of numbers within a tribe. Two pairs of cousins many, making

-two new fire-circles, and have, say, two children apiece. These

EXOGAMY AND THE MATING OF COUSINS

also marry. The result is trvvo family-circles and perhaps four
cliildreii, who may in their turn marry. If cousins had been for-
bidden to marry, we should have had eight fire-eirclea and perhaps
sixteen cliildren. Exogamy thus in the wider sense, but not
McLennau's, has an important bearing on the making of nations.
In such a dual family as we are assunciing, it will soon happen that
the supply of cousins fails, the balance of the sexes will be unequal ;
young men will therefore have to get their wives from elsewhere,
or young men from elsewhere may be allowed to join the group.
It is not likely that this latter method of getting rid of superfluous
women would be adopted at an early stage, polygamy would be
preferred. But polygamy seems a rather late development, and in
any case there would be a limit to the polygamous capacity of early
man ; male individuaUsm, moreover, would object to male intruders.
However, allowing for these exceptions, the main point is that
sometimes a man would get a wife from a friendly group, by
exchange of a sister or other arrangement. It is just here that
* capture ' of wives was supposed to come in. I do not deny that
such capture may occasionally have occurred, but I hope I have
elsewhere shown that the hypothesis of a period in human history
in which * marriage by capture ' was an institution has no foundation.
It is of the rarest occurrence in Australia, for such acts lead to war,
and early man is not fond of such disturbances. He is perforce as
peaceable and harmless as may be ; in this Mr< Payne agrees. The
capture of women in time of war is a very different thing.

It is this introduction of fresh women that brings new blood
into the family, and causes it to expand by producing new branches
of the two original families, in time raising the dual family to the
proportions of a tribe.

My suggestion as to the origin of dual exogamy is con-
firmed by the following. Dr. Codrington says of the Melanesians,
' in the native view of mankind, almost everywhere in the islands,
nothing seems more fundamental than the division of the people
into two or more classes, which are exogamous, and in which descent
is counted through the mother. ... No single family of natives can
fail to consist of members of more than one division/ The same
two divisions run through the Banks* Islands, with the Torres
Islands and the Northern New Hebrides. In neither the Banks'
nor the New Hebrides is there a name to distinguish the division or
kindred ; nor is there any badge or emblem belonging to either ;

EXOGAMY AND THE MATING OF COUSINS 59

' in their small communitm every neighbour is well knounL Each of the
divisions is in Mota called a vevCy in Motlav tei^, a word which in
itself signifies *' division ". Those who are of one veve are said to be
tamla ma to the others, i.e. *'of the other side of the house.*'
A woman who marries does not come over to her husband's side of
the house ; she is said to be ape mateima^ ** at the door/' All of the
same side of the house are sogoi to one another. Hence a man's
children are not his sogoi, ** his kindred," liis nearest relatives are his
sisters children. Within the two veve there are certain families
among the Banks* Islands people/ In Aurora, and Maewo of the
New Hebrides, the members of the two divisions speak of one
another as ' of the other side *, la tavuluna. Several families are
found within the kin ; most are named from places, one from the
octopus, but these have no notion of descent from it, and eat it
freely* To these family groups the same name veve is given as to
the two great kindreds. In Lepers' Island the two divisions are
called * bunches of fruit ', Wai vung^ as if all the members hung on
the same stalk.

The Fijian classificatory system is as follows : —

4. my Sister
UuBbaud.

3. my Sister
Huslmnd.

1. Ego (male)
Wife.
\

2. my Brother
Wife.

5. Son. 6, Dau. 7. Son* 8. Dau. 9, Son. 10. Dau. 11. Son, 12. Dau.

5 and 7
9 and I ]

6 and 8

are respectively veitathinL

10 and 12
{iathi is the term of relationship between brother and brother or sister and
f«t8terO

5 and 7 are vemgaiteni with 6 and B.

9 and 11

10 and 12.

one

{n^ime is the term of relationship between brother and sister. It means
. who shuns *, and the veinffaneni are the non-marriageable persons.)
5 and 7 are veimlavolani with 10 and 12*
9 and 11 ,. ,, ,, 6 and 8.

(veindavolani are the only marriageable persons and are expected to marry.
Davola means tnarria^eabk.)

Here the germ of two pliratries is evidently the cross-cousin-
marriage. (It is to be noted that there is evidently an etymological
eonnexion between the words tavala of the Banks' Islands, iavuluna
of the New Hebrides, and the Fijian davola.) The veindavolani are
* of the other side of the house * and are marriageable. Apply family

EXOGAMY AOT) THE MATING OF COUSINS

names to these and we have two phratries, continually repeated by
cross-coiisin-marriage.

We have mentioned that the families of one phratry of the
Iroquois are ' cousin^families ' to those of the other. The Karens of
South Burmah, says Mason, * have two principal divisions, the Sgaus
and PwoSj which are indicated as Pah4ee (** of descent from the father's
side ") and Mo-tee (** of descent from the mother's side ")/

The Chinese have one set of terms for the ancestors on the
fathers side, and another for those on the mother's. The Zulus
mark the distinction in the same way for the first set of
ascendants.

I think this theory of the origin of the two-phratry system^ may
daim the advantages that (1) it explains the bisection as a natural
growth without calling in the aid of any arbitrary and deUberate
legislation. Here I am at one with Mr. Lang. It gives a method
by which the division could arise automatically ; (2) it explains (and
these are difficulties in other explanations) why the families of one
phratry may not marry among themselves; (3) it does not begin
with local exogamy; (4) it enables us to do without the self-
contradictory and unwarranted hypothesis of an * undivided com-
mune ' with aU its difficulties, especially the difficulties of getting
into it and of getting out of it. Here again I agree with Mr. I^ang.
Later on it will produce another argimient against * group^marriage '
soKsalled ; (5) it coincides with the express statements of all those
aboriginal thinkers (whose wits are not inferior to those of the
average 'civilized* man), to the eflfect that all these exogamous
groupings are connected with kinship, real kinship, though con-
veniently, as with us to some extent, identified with name-
kinship ; (6) it excludes from an unwarranted pre-eminence
the system of totemism; (7) it is of universal application* It
explains those rare cases where the phratries are more than two ;
those where they exist, but have no names, mere * sides of the
house ', and those in which various names, sometimes fanciful, have
been applied later. It shows that the germ of dual exogamy is
contained in every marriage and therefore in every family ; those
peoples who have not developed this, have to thank better circum-
stances, less external pressure, than fell to the lot of people like the
tralians; (8) and lastly, it enables us to trace the origin and
ih of the tribe, in a natural and convincing way, from the
ly. Mr. M'(jree, of the Bureau of Ethnology, states, as the

EXOGAMY AND THE MATING OP COUSINS 61

accepted result of American research, that some small discrete
group, probably the family, was the earliest unit. Mr. E. J. Payne,
in his History of tlw New WorMj is of opinion that the tribe can
originate from the family. Indeed, what else is it to originate
from? It would seem that some few peoples thus retain, not as
a mere survival, but as a living institution, the actual machinery by
which the process was carried out. I may here note that Mr, Lang's
original local groups consisting of several family-names, seems too
large not only for Mr. M*Gee*s * small discrete bodies, probably
family-groups ', but for the patriarchal jealougy of the polygamous
ape of Mr, Atkinson. I should like to add, as bearing on connected
issues, that Mr, M'Gee is of opinion that those families ' must have
been essentially, and were perhaps strictly, monogamous ', and that
he scouts the postulates that the primitive Americans were arrayed
in ' chaotic hordes, and that organized society was developed out of
this by the segregation of groups ' (as Mr. Howitt seems to hold),
and ' that the primal conjugal condition was one of promiscuity \ If
we need an analogy fi*om some higher animal for the primitive state
of man, connubial and social, we really cannot bring forward the
baboon.

I would therefore place the order of development thus : the
family, combination of two families (resulting sometimes, owing to
external pressure, in a tribe with dual exogamy), or of more than
two ; then further combination ; tribal status can be assumed at any
time, war would organize the related families. Gentes, gotras, clans,
thums, septs, and phratries are all practicaUy identical, they are
families with family names.

I noted in The Mystic Rose that the Arunta system resulted in
the prevention of marriage between first cousins. Among the
Karens first cousins may marry but are thought too near ; the most
suitable match is that of second cousins ; third cousins are thought
too remote, and beyond this relationship marriage is forbidden.
Tlius with the Northern Arunta^ the original ' age-class ' system has
ended by preventing what no other classificatory exogamy can of
itself prevent, the marriage of first cousins.

The difference of name, combined with the intermarriage of the
two families, makes all ^ s and J5 s of the same generation marriage-
able ; hence they are conveniently termed * husbands * and * wives \
In the Ta-ta-thi tribes with two phratries, every Mukwara man
speaks of every Kilpara woman as * wife ' and vice versa. The

EXOGAMY AND THE MATING OF COUSINS

members of the younger generation in a tribe are thus always
*brothei-s and sisters' or * husbands and wives*. Now without
exception, so far as I know^ it is the ease that the real relationship
is known, as Mr. Howitt says of the Kumai, ' every one seemed to
be the father, mother, son, daughter, brother and sister of every
one else, but when special inquiry was made, the ** tribal " relation-
ship was distinguished from the ** own " by more precise statement,
aa " the other father", " other mother,** &c/ So Dr. Codrington says
of the Melanesians that all of the gi*ade above the * brothers and
sisters' are * fathers and mothers', but they distinguish 'own
fathers * and ' own mothers *. The same is the case with the
Fijians who have no phratry names, but terms of relationship,
veinganeni for the 'brothers and sisters', veindamhni for cousina
As to the relationship of ' husband * and * wife ', Dr. Codrington
says * to a Melanesian man all women of liis own generation are
either ''sisters" or "wives", to the Melanesian woman all men
are either '^ brothers" or "husbands'*. It must not be understood
that a Melanesian regards all women who are not of his
division m in fact his wives^ or conceives himself to have rights
which he may exercise in regard to those of them who are
unmarried/

The existence of the two sides of the house is thus the key to
classificatory relationship. There are one or two other results. This
explanation enables us to account for these ' tribal ' or ' group-
relationships ' such as ' brother and sister ', ' husband * and ' ^vife *,
* father * and ' mother \ without calling in the aid of ' group-
marriage ' or promiscuity. Hr. Cunow has made on this foundation
an unanswerable argument against the * promiscuity ' theory. We
note also that this system, while extending terms like ' brother ' and
' father ' beyond their natural meaning, ignores the relationship
between the two sides of the house. It is assisted in this by the
difference of family-name, and by the principle that man and wife
do not become akin by marriage.

From what has gone before, it is evident that the members of
a typical exogamous tribe are all closely related; how far they
themselves recognize the fact does not matter ; but the fact has an
interesting bearing on the formation and persistence of racial type
and character. Morgan noted this, but his explanation of the close
kinship was wrong.

Preservation of type, reversion to type, variation from type, are

EXOGAMY AND THE MATING OP COUSINS 68

among the most interesting of biological phenomena. The possibility
that oousin-marriage has acted for long ages as a counteracting influ-
ence against variation opens out the way to curious reflections. The
variational tendency, on the one hand, is connected, biologically, with
out^breeding, and psychologically with romance ; herein is progress.
Nature, so careful of the iype, is assisted by the conservatism of man,
his solidarity even, to keep the balance. Perhaps the race owes
more to family-alliances than it wots of.

THE AUSTRALIAN FOREHEAD

By D. J, CUNNINGHAM, M.D., D.Sa, LL,U, D.C.L., RRS.,
Professor op Anatomy, University of Edinburgh.

When a large series of Australian crania is examined we are
presented with an assortment of diverse characters of a somewhat
extreme kind. HiisJey, in his classical essay upon human fossils (1),
alludes to the differences that are met with in respect to cranial
height. * Many Australian skulls,' he says, ' have a considerable
height, quite equal to that of the average of any other race, but there
are others in which the cranial roof becomes remarkedly depressed/
Still more obvious are the differences which may be noted in the
r^on of the forehead, because these are more easily appreciated by
the eye and only require the application of the more refined cranio-
metrical methods when we seek to estabUsh the precise degrees of
difference in this respect.

The forehead may be defined as that region of the cranium
which lies above the nasion and orbits and below the coronal suture.
On either side it is bounded by the fore part of the temporal ridge.
Its limits are thus very definite. It consists of two parts, a lower
glabellar and supraorbital part, the region of tlie eyebrows, and an
upper cerebral portion. I purpose excluding the former from the
scope of this paper ; not because it fails in interest — ^indeed it is the
more interesting of the two districts, and presents more marked
differences within the limits of this race than perhaps can be
obsen^ed in any other race — but simply because the region is so
important that it requires separate and special treatment, and tliis
I hope te give to it in a future paper.

As a rule the cerebral part of the forehead in the Australian is
not flat and receding, and in many cases it presents a curvature as
bold and pronounced as that which characterizes the European. But
in all large collections of AustraUan crania specimens will be found
which present a degree of frontal flattening which is met with only
in higher races in rare cases and then chiefly in microcephahc or
deformed crania.

THE AUSTRALIAN FOREHEAD

In the ethnological section of the Museum of the University of
Edinburgh, which contains, through the influence of Sir WilUam
Turner, and the generous donations of many graduates, of whom may
be specially mentioned Dr. W. Ramsay Smith, a unique collection of
Australian crania, there are many specimens from different districts
of Australia which show thLs depressed and degraded type of
forehead.

It is necessary that we should devise some method of measuring
this cranial character in order that we may give to it its true and
appropriate value. Schwalbe (3 and 4) has pointed out that the
frontal inclination depends upon two quite distinct factors, viz.

(1) the degree of elevation or depression of the frontal bone and

(2) the degree of cur\'ature exhibited by the bone; and he has
adopted a procedure which he believes affords information on both
of these points.

The views which he entertains regarding the evolution of the
cranium of recent man out of the type presented by the palaeolithic
remains of Neanderthal and Spy, have led this distinguished
anatomist to attach too much importance to a supposed elevation of
the frontal bone, by means of which its upper border moves upwards
and forwards. The diagram which he employs to illustrate his views
on this point, and which we have taken the liberty of reproducing
(pL V, fig. 1), would ahnost seem to indicate that he believes the
change to be brought about by a process of rotation of the bone
around a transverse axis drawn through the nasion.

By this movement of the frontal bone Schwalbe believes that
the bregma is displaced forwards and that the whole bone becomes
more vertical. To estimate the extent of the change he employs
three methods : —

1. The determination of a frontal angle.

2. The determination of a soK^alled bregma angle,
8, The determination of the position of the bregma.

A necessary preUminary to the employment of each of these
methods consists in obtaining an accurate contour tracing of the
mesial longitudinal arc of the cranium. This can be done with
great exactitude by Lissauer*s Diagraph. Upon the tracing thus
acquired Schwalbe draws a base-line from the inion to the most
prominent point of the glabella.

To determine the frontal angle a line is drawn from the anterior
end of the base-line so as to touch tangentially the most projecting

THE AUSTRALIAN FOREHEAD

•

point of the curvature of the frontal bone. The angle which this
forms with the glabella-inion line constitutes the angle in question.

The hreffma angle is obtained by carrying a line from the gla-
bellar end of the base'line to the bregma and measuring the angle
which it forms with the base-Une ; whilst an endeavour is made
to ascertain the position of the bregma by dropping a perpendicular
from this point upon the base-line and calculating the relative
nearness to or distance from the glabella of the point of intersection.
Each of these three methods is more or less faulty and liable to
lead us into serious error.

Two discordant factors determine the degree of acuteness of
the frontal angle, viz. the slope of the forehead and the amount
of projection exhibited by the glabella. In the Neanderthal cranium,
where the glabella is enormous, the lower end of the frontal hne
is thrust far forward, and the angle expresses this character quite as
much as, if indeed not more than, the frontal inclination* It would
be useless therefore to compare the frontal angle in such a cranium
with the same angle taken, say, in an Andaman Islander, where
the glabella is almost inappreciable as an eminence. For the same
reason we cannot employ it for the determination of the degree
of frontal elevation even in a single racial group such as the Austra-
lians. Amongst the crania of this race we meet with every kind
of glabella, from one almost as excessively developed as in the
Neanderthal cranium to one wliich is not more elevated than in
the ordinary European skull.

The bregma angle suffers from even greater disabilities, because
not only is the lower end of the bregma hne subject to displacements
due to factors quite outside the slope of the frontal bone (i. e. varia-
tions in the glabella) but also the upper end is subject to changes
in its relative position on the cranial vault quite independent of
those which may arise from an elevation or depression of the frontal
plate* It is not necessary, therefore, to discuss the results obtained
by the Strasburg anatomist by these two angles.

But still further: the different positions of the bregma in
different skulls ascertained by dropping a perpendicular from that
point on to the base-line does not necessarily indicate a growth move-
ment of the frontal bone leading to an elevation or depression
of the bone. In the phylogenetic evolution of the forehead of
recent man there has been, as every one must see, a general tendency
towards the expansion and extension of the frontal district, and

F 2

THE AUSTRALIAN FOREHEAD

also concurrent with this an elevation of the bregma as we pass
from the lower to the higher types ; but it does not follow from
this that the bregma has shifted its original relative position on the
cranial vault to any great extent, or that what shifting there is can be
accounted for by a lifting up of the posterior border of the bone.

The diflFerent positions occupied on the base-line by Schwalbe's
bregma perpendicular can equally well be explained (1) by differ-
ences in the degree of development of the glabella, (2) different
degrees of growth-extension of the frontal area of the cranial
vault, and (3) different degrees of growth-extension of Uie parietal
and occipital portions of the cranial wall affecting the length of
the base-Une and the level of its posterior end. To discuss at the
present moment the interesting evolutionary increase in the length
of the upper margin of the parietal bone hinted at by Huxley and
so fully and clearly established by Schwalbe and the still greater
extension of the squamous part of the occipital bone in recent man
would open up too large a question, and could not be properly
discussed within the scope of this paper, but I thoroughly realize
the importance of this factor in connexion with the present con-
tention. I have no desire to evade the difficulty (if indeed we can
call it a difficulty), and it is my intention to deal with this aspect
of the evolution of the human ci^anium at no distant date.

It is the common practice of craniologists to measure and
compare the relative extents to which the frontal, parietal, and
occipital elements enter into the constitution of the mesial longitu-
dinal arc of the cranium. No one has brought out the differences
met with in this respect more clearly than Sir William Turner
(6, 7, and 8) in his numerous important memoirs on the craniology
of different races. From these and other writings it may be seen
that the relative mesial length of the frontal bone is not always
the same in different races, nor indeed in different individuals of
the same race. Still it should be noted that where such investiga-
tions lead one to infer a shifting backwards or forwards of the
bregma the real change may be in the parietal and occipital elements
of the cranial arch^ whilst the frontal element may have to a large
extent remained unaltered.

Still, the presumption is that there is a certain amount of
variabiUty in the position of the bregma on the cranial vault of
recent man due to an increase or a diminution of the frontal district
of the cranial vault, and this constitutes another disturbing element,

THE AUSTEALIAN FOREHEAD

quite outside the question of the possible elevation or depression of
the frontal bone, which tends to vitiate the results obtained by
Schwalbe's bregma angle and likewise the conclusions which he
draws from the position of the bregma ascertained by his bregma^
perpendicular.

Schwalbe (3) fully realizes the influence which the length of the
frontal bone exerts on the index which he constructs upon the
position of the bregma perpendicular on the base-line. He specially
refers to the high index in the New World ape, due, not to a depres-
sion of the frontal bone, but to its great length and also to the low
index in the orang due to the shortness of the frontal bone.

An effort to obtain some precise and definite evidence regarding
the limits of variation in the position of the bregma on the cranial
vault afforded some interesting information. The method pursued
was to take a similar point, wliich I shall call the ^ third point \ on
aU the crania examined, and determine the position of the bregma in
relation to it.

The point in question is obtained by taking the place of junction
of the anterior and middle thirds of the mesial longitudinal arc
measured from the nasion to the opisthion. The measurements for
its determination in a large number of skulls of different races are
ready to hand in Sir William Turner s numerous important memoirs.
Those are the figures which I have used in connexion with this part
of the inquiry.

The * third point ' generally lies very close to, and often coin-
cident with, the bregma. It is, as a rule, a few millimetres in front
of, and very rarely behind, tlie bregma. Its mean position in five
Fuegian and Patagonian skulls is one millimetre behind the bregma,
and in eleven skulls of Admiralty Islanders (six males and five females)
its mean i>osition is coincident with the bregma, and in the females
1-2 mm. behind it. These are exceptional cases. In all the other
races which have been studied from this point of view its average
position is always in front of the bregma, although individual cases
are met with in wiiich it lies from one to five millimetres behind it.

The following table gives the results : —

THE AUSTRALIAN FOREHEAD

TABLE I

POBITION OF THE 'THIRD POINT* WITH REFERENCE TO THE BrEOICA ASCER-
TAINED FROM THE MEASUREMENTS GIVEN IN SiR WiLLlAM TuRNEB's

Memoirs.

Average distance in nun. in/nmt of the Bregma.

Highest degree of

variation.

Bushmen

Ko.

Males

Females.

Sexes mixed.

7.4

13 mm.

Australians

3-4

■ •1

16 „

Sandwich 1
Islanders j

4-7

...

19 „

Chatham )
Islanders j

3-6

...

18 „

Hftoris

&-4

...

21 „

Yeddahs

6-5

18 „

Chins

...

**•

4-7

16 „

Tamil Sudras

8-4

■ <•

1* ,.

Burmese

...

• •»

6-7

17 „

Chinese

• ••

• •»

4.1

14 „

Scottish

...

26 „

The mean distance of the * third point ' in front of the bregma
in the male varies in diflferent races between the limits of 6 mm.
and 8 mm, ; in the female the variation is not so great, a circum-
stance which may be partly accounted for by racial differences in the
prominence of the glabella.

The limits of variability are seen to be considerable. Thus in
the Scottish skull there is a section of the cranial arc 26 mm, long, on
any point of which the bregma may be placed. The details in regard
to the Scottish skulls may be given in somewhat fuller detail.

TABLE II

Scottish ekxjll& grouped accorbiho to position of the

RELATION TO THE BrEQMA.

Hales.

Group L * Third point* 11 to 21 mm* in front o/ Bregma

Group 2* * Third point * 6 to 10 mm. infivnt of Bregma

Group 3. * Third point * 1 to 5 mm. in front q/ Bregma

Group 4. * Third point * caincidefd with Bregma

Group 6. ^ Third point * 1 to 5 mm. behind the Bregma

Females.
Group 1. * Third point' 12 mm. in front o/ Bregma
Group 2. ' Third point ' 6 to 1€ mm. in front o/ Bregma
Group 3. * Third point' 1 to 5 mm. in front o/ Bregma
Group 4. * Third point * coincident with Bregma
Group 6. * Third point * 1 to 5 mm. behind the Bregma

THIRD POINT IN

^0. qfgkuUs.

1
15
18

THE AUSTRALIAN FOREHEAD fl

A study of these tables brings out an interesting sexual
character in the position of the bregma. In all the racial groups
in which it was possible to diflferentiate tlie sexes the bregma is
situated relatively further forward on the vault of the cranium
than in the males. To some extent this result is no doubt due to
the higher degree of development of tlie glabella in the male causing
an increase in the frontal measurement ; but I am satisfied from a
study of the influence of the glabella in this direction that the whole
difference cannot be accounted for in this way, and that there is a
real sexual distinction to be noted in the position of the bregma upon
the mesial longitudinal arc of the cranium.

Tlie question naturally arises : Are the different positions of
the bregma which we have noted associated in any way with the
many forms of cranium which distinguish the racial groups in which
the observation has been made ? So far I have failed to determine
any such correlation. The mean amount of variation as well as the
limits of individual variation are very much the same in the lofty
crania of the Sandwich Islanders as in the more depressed cz^ania of
the Australians. The height index would therefore seem to be in
no way correlated with the position of the bregma* Nor does the
cephaUc index appear to be any more closely associated with it. In
the brachycephahc Oahus of the Sandwich Isles the bregma is placed
4 mm. behind the * third point ' ; in the dolichocephahc members of
this group 4-3 mm, behind it.

Schwalbe s view that in the phylogenetie development of the
human skull there has been a process of elevation of the frontal
and occipital elements and that the cranium therefore opens out
like the bursting of a bud is an attractive and ingenious conception.
The evidence on which it is based, however, is not in every respect
satisfactory.

The percentage which each of the elements of the cranial vault
contributes to the mesial longitudinal arc in recent man and an ape
is very instructive.

Scotch CraDiam of

Cranium. Macaquti Monkej.

Frontal 332 iSB

Parietal . . 85-5 834

Squamous part of CMjcipital .... 214 5»7

Nuchal part of occipital , . . • . 10-2 17-6

100- 100'

These figures show that before we can decide upon the extent

THE AUSTRALIAN FOREHEAD

to which the lifting up of the frontal and occipital elements takes
place there are niany other matters which require explanation.

At the same time I am far from asserting that in recent races
the form of the forehead is entirely due to growth changes which
affect the curvature of the frontal bone and the extent of the area
it occupies in the cranial vault* There is evidence which seems to
indicate that the frontal plate as a whole may be more elevated
or more depressed in certain individuals and in certain races than
in others ; but this plays a minor pait in determining the forehead
contour.

To establish tliis point it is necessary to discard Schwalbe's
base-line and to replace it by one which extends from the inion
to the nasion. The latter point is as fixed and as constant as we
can expect any such point to be on a structure such as the skull in
which there are so many fluctuating influences affecting its growth
and form. The inion is not so satisfactory^ because its position is
certainly subject to a certain amount of variation. It should be
noted that we employ the tenn inion in the sense in which it was
employed by Broca, and in which it in at present used by English
anatomists.

Schwalbe himself recognizes the superior advantages of the
nasion-inion base^liiie, and it is difficult to understand why he has
selected the glabella-inion line in his study of the Neanderthal
cranium seeing that in it as well as in one of the two Spy crania
the nasion is preserved.

We have observed that the extreme degree of variability
in the position of the bregma on the cranial arc in the ninety
Scottish crania examined is 26 mm. If this be measured on a
contour-tracing of the mesial cranial arc so that the bregma lies
exactly in the middle and two lines drawn from either end of tliis
portion of the arc to the nasion two angles are formed with the
base-line which measure respectively 53" and Q2\ At least these
were the results obtained in an Australian contour-tracing which
I selected at random for this experiment (pL v, fig. 2), The magni-
tude of these angles would obviously be aSected by the height of
the cranial vault ; in a higher cranium they would be reduced ; in
a lower they would be increased ; but this does not aflfect the result
that in an average skuU of the group we are dealing with the two
extreme positions which may be assumed by the bregma yield a
difference of only 9' in the bregma-nasion-inion angle.

THE AUSTRALIAN FOREHEAD

But in the twenty Australian skulls taken from Sir William
Turner's lists the amount of variation in the position of the bregma
was only 16 mm. Now if we had two skulls of equal height
which exhibited the two extremes of this degree of variability the
difference of the bregma-nasion-inion angles would not be more
than B\ As we shall see later, this angle in the Australian skulk
varies between the limits of 52^ and 65*, which gives a range of
variability of 13',

From this it may be inferred that an elevation and depression
of the frontal plate in recent man does take place to a small extent
and has to be reckoned with as a factor which in some degree
influences the form of the forehead.

The BregmorNaskm-Inion or B. N, L Angle (pL vi, fig, 3),

In the ethnological section of the Museum of the Edinbui^h
University there are more than 100 skulls of Australian natives. The
large amount of time which is required to obtain accurate contour-
tracings of the mesial longitudinal arc of the cranium and the
pressure of other duties have rendered it impossible for me to
undertake the examination of the whole collection. I have therefore
selected two groups, viz. thirteen skulls from Victoria and twelve
from Queensland, and to these I have added three which presented
a high degree of forehead flattening. These latter comprise a male
skull of the Milang tribe, South Australia, specially referred to by
Sir William Turner in his Chalknger Report (p. 46), a female skuU
from Central Australia which exhibited syphilitic (?) disease of the
calvaria, and a male skull from New South Wales.

The size of the bregma-nasion-inion or B.N. I. angle in these
twenty-eight skulls is given in the following table : —

THE AUSTRALIAN FOREHEAD

o «

^ coO

I *-4

.. o

•^ C^ #3<s
"» * 0^1-1

S i§§S

o§2

?.ss

s»

1^3

1^1

i^i

•I

-< 1,00

> 10 (
CO

c9o^
«co»9

IC<I

5 i
g £

!*H

<f>eoi-ico
I

1 ^
1^

i'l'

%'i'

Sill

, ^"?

61°

208
184°

^^§1

60°
-7
21-8
185°

fetal

S7g3|

d c fc fc
^-3 5 5

5 6.

^ tH tH

^rs

1 i

o CO®-

^^^^^^ THE AUSTRALIAN FOREHEAD 76 ^^H

The B. N. I. angle varies from 52"* to 64°, The fact that twenty-one ^^M
of the twenty-eight skulls examined presented an angle which ranged ^^M
between 57' and 62"* ^hows that there is a considerable degree of ^^H
constancy in this respect and that the extreme conditions, both high ^^M
and low, are not particularly common. ^^M
It will be seen from the table that the mean angle for the males ^^M
(excluding the three specimens specially selected on account of their ^^M
low-forehead formation) is 604'. The females show a slightly lower ^^M
angle, but it is possible that this difference might not be maintained ^^M
' if a larger number of specimens were measured. Still we should ^^H
not lose sight of the fact that the bregma is placed relatively further ^^|
back in the female than in the male, and that the more acute B. N. I, ^^|
angle in the former may be due to this. ^^H
^H For purposes of comparison I took contour-tracings and ^^|
measured the B. N» I. angle of eight Scottish skulls. Seven of these ^^|
were taken more or less at random, although it is right to state that ^H
I tried to include in the series one or two specimens which seemed i^^l
to be more flattened in the forehead region than the others. Six of '^^^|
the selected skulls were males and one that of a female. The eighth, ^^|
that of a male, and known as the Aberdeen skull, was included on ^^|
account of its exceptionally low and degraded type of forehead. It ^^H
was found during digging operations on the site of the old Blackfriars ^^|
Monastery in Aberdeen and was described many years ago by Sir ^^|
WUham Turner (9). ^H

^K TABLE I¥ ^H

^^^^^ SconriSH Crania (6 male, 1 female). ^^H

^^^* B, N, I. angle and tbe Index of Frontal Curvature. ^^H

No. offikuIJ.

2 '

7
21*7

Mean resulto.

B. N. I. angle ,
Index of the Frontal \
Curve J

60^
20-2

60°
262

60" 1
22-1

254

28-8

62"
252

615°
23.7

In the above table the B. N, L angle is one degree higher thai
that obtained for the AustraHan male, and my belief is that thi
[iiflference is a real and actual one. Still it must be noted that th<
liigher mean obtained for the Scottish skulls depends upon th<
exceptionally large angle present in one specimen {No. 7).

In the examination of the Scottish skulls it became apparen
that the dififerences in the angle were as much due to differences ii

THE AUSTRALIAN FOREHEAD

the level of the inion as to differences in the position of the bregma
or of the degree of elevation of the frontal bone. Should this
observation prove correct the stronger musculature in the neck of
the Austrahan and the concurrent slightly liigher relative position of
the inion would account for any difference there may be in the
Australian and Scottish B. N. I, angle. This is a matter which
requires further investigation.

Tlie Aberdeen cranium presented a B. N, 1 angle of 53' (pi. vi,
fig. 4). Tliis is quite exceptional, but it shows that amongst Scottish
skulls an angle as low as that in any Australian cranium may be
encountered.

Degree of Fro^ital Curvature.

The predominant factor which determines the verticaUty or
depression of the forehead is the degree of curvature or bulge of
the frontal bone. This cranial character can be measured with
considerable exactitude. Schwalbe employs two methods.

1. Lmatiers method (2). Two lines are drawn from the point
of highest convexity of the cun^e to the two extremities of the
frontal chord (nasion and bregma) (pi vi, fig. 3, a.n. and a.b.) and
the angle which these enclose is then measured. Schwalbe terms
this the angle of the frontal cur\^ature,

2. By measuring with a tape the mesial arc of the frontal
bone from the nasion to the bregma and comparing the result
with the length of the chord measured by tlie callipers between
the same two points —

Frontal chord x 100
Frontal arc length.

Lissauer's method requires a contour-tracing of the mesial
arc of the cranium. It yields accurate results. The more open
the angle the flatter the curve, and vice verm* According to
Lissauer the angle is 171 in the goriUa and 120** in the negro.
He also points out that in young skulls there is always a higher
degree of curvature than in the adult.

The second method has as a disturbing element the varying
degrees of projeetion of the glabella ; in cases where the glabella
is high the index exaggerates the degree of curvature and even
in skulls with a low glabella the condition is not accurately expressed
by the index.

THE AUSTRALIAN FOREHEAD 77

There is, however, a simpler plan than that suggested by Lissauer,
and one which is easier of application^ less liable to errors arising
from the nianipulations involved, and which expresses the result
in a more graphic and intelligible way. The height or degree of
curvature of the calvaria is determined by dropping a perpendicular
on the base-line (nasio-inial line) from the highest point of the
mesial longitudinal cranial arc and comparing its length %vith that
of the base-line^ —

Calvaria height x 100
Length of nasio-inial base-line.

The same method may be apphed to determine the degree of
curvature of the several segments of which the mesial cranial arch
is composed {pL vi, fig. 3). The index of the frontal curve or frontal
height may be ascertained thus : —

Height of Frontal Curve x 100

Length of Frontal Chord (nasio-bregma line).

In the table given on p. 74 the angle of the frontal curvature
and the index of the frontal curve are given for the twenty-eight
Austrian crania which have been studied.

The average frontal curvature angle for the male skulls from
Victoria and Queensland is 133" ; in the female skulls of the same
groups it is somewhat smaller, thereby indicating a higher degree of
cur%^ature. We can accept this as a sexual character.

In Schwalbe's memoir on Pithecanthropus the following
measurements of the frontal curvature angle in different races are
given : —

Negroes {10) . 125 6**

Male Natives of Alsace (24) ....•*, iai3*»
Kalmucks (4) 1369^

It is interesting to note that the negro, according to these
figures, has a stronger degree of frontal curvature than the European.
There can be little doubt that the feature is one of some racial
importance. The Australian presents a mean angle only sliglitly
more open than that of the native of Alsace* At the same time it
should be borne in mind that amongst the Australians an extreme
degree of flattening of the forehead occurs with a considerable degree
of frequency. In the twenty-eight skulls examined there were four
specimens with an angle which varied from 140' to 146 ^ Three of

THE AUSTRALIAN FOREHEAD

these were not included in the calculation which afforded the mean
result. It is indeed a matter for surprise that in three of the Alsatian
skulls measured by Schwalbe the angle was 140 ^ 140 5*, and 143'.
Amongst a veiy much larger number of Scottish crania only
one, the Aberdeen cranium, had an equivalent degree of frontal
flattening, and it has always been regarded as a unique specimen in
this respect.

As I have said, I am inclined to place more reliance on the
results yielded by the index of the fi'ontal curve than on the angle
of the curve. A glance at Table III will show tliat the results
obtained by the two methods correspond very closely ; still there are
slight discrepancies, and these I attribute to the difficulty attached to
the estimation with absolute accuracy of an angle so open.

The mean index of the frontal cur^^e in the male natives of
Victoria was 22 4 and for the male Queenslanders 21 '4. The
mean index for the latter was reduced by the inclusion in this
group of one skull with a very high degree of frontal flattening
(Ko. 6).

The higher degree of frontal curvature in the ScottLsh (see
Table IV) is indicated by a mean index of 23 7, although in the
exceptional Aberdeen cranium the index fell so low as 15*9.

We have already referred to the higher degree of frontal curva-
ture in the Australian female (mean index 23-7) than in the Austra-
lian male (mean index 218). This sexual character is also brought
out by Schwalbe, who gives the angle of the frontal curvature for the
female Alsatians as 129 6"" and for the males as 131*3\

There is little or no relationship between the B, N. L angle and
the index of frontal curvature ; thus a low^ angle (56'') may be associated
with a high degree of curvature (index of 25 4), or the same angle in
two different skulls (60 } may be correlated with curvature indices of
such different values as 13*2 and 26*2. Still, in cases where the
angle falls below 65" the index of curvature is, as a rule, also low, e,g»
skull of Milang tribe, the female skull from Central Austraha, the male
skuU from New South Wales and the Aberdeen cranium (pi. vii).

The frontal bosses are usually very feebly developed in the
Australian skull, and in many cases they are actually absent as
appreciable eminences. A median frontal ridge is very common in
the male. It was present in most of the Victorian males ; in the
Queensland skulls, on the other hand, it was, as a rule, absent ; in
fact, in only two specimens could it be said to be present.

THE AUSTRALIAN FOREHEAD

EXPLANATION OF PLATES V-VII

PLATE V

Via, 1. From Schwalbe's Memoir on Pithecanthropua (3)» Diagram to show
the different positions of the bregma as the frontal bone is raised or depressed^
and also to illustrate Schwalhe s method of estimating the extent of these changes.
The diagram has evidently been constructed from the con tour- tracing of one
skull, and the ypper and lower dotted outlines of the frontal bone have been
obtained by rotating this upwards and then downwai^s around the nasion as
a centre,

Fia. 2. Mesial contour-tracing of the fore-part of an Australian cranium.
Two points, 26 mm. distant from each other, are marked on the cranial vault,
with the bregma (b) between and equidistant from each. To each of these points
a line is drawn from the nasion (n) and the angle in each case measured. This
may be considered to give approximately the limits of variation in the B.N.L
angle due to variability in the position of the bregma,

PLATE VI

Fio. 3. Mesial contour-tracing of the cranium of a female Australian.
I.V. Baae-line* k.b. Kasion-bregma line or the frontal chord. b.k»i. Bregma-
nasion-inion angle, a. Highest point of frontal curvature. b.a.k, Lissauer's angle
of the frontal curvature.

Fio. 4» Mesial contour-tracings of the fore-parts of three skuUs, which show
a very different conformation in the forehead region. In the upper Scottish skull
the bregma is placed unusually far forwards and the frontal bone presents a high
degree of convexity ; in the lower Scottish skull (the Aberdeen cranium) the fore-
head is flat and depressed. The common baae-line upon which the tracings rest
pftsaed through the inion and nasion in each case.

PLATE VII

Fio. 5. A series of mesial contour-tracings of the forehead arranged on a
common base-line which extended through the nasion and the inion in each case*
The B.N.I, angle is given and the tracings have been selected with the view of
showing a regular gradation according to the magnitude of this angle.

With the exception of the last five tracings on the lower base-line, all the
tracings have been taken from Australian skulls, and further particulars regarding
each may be obtained by referring to Table III, p. 74 of the text

y* indicates a Victorian skull ; and 9, a Queensland skull ; and the numbers
associated with these letters give the Museum numbers of the specimens. n.s,w. is
the New South Wales skull ; Milano, the Milang skull ; and c.a. the female skuU
from Central Australia. The three last specimens are referred to in the lower part
of Table ni.

M., JOE, B.L., and H.S. are frontal tracings from the crania of four micro-
cephalic idiots.

Gob, is the frontal tracing of a young female gorilla, the skull of which is in
the Anatomical Museum of Trinity College, Dublin.

80 THE AUSTRALIAN FOREHEAD

LITERATURE REFERRED TO IN THE TEXT

1. HuxLET, T. H.— On Some Fosail Benudns of Man. CoUeded Essaifs, VoL vii,

p. 167, 1894.

2. LissAUBB. — Untersuohungen Qber die sagitiale KrOmmung des ScbftdeLs bei

den Anthropoiden and den verschiedenen MenBchenrassen. Arckw fUr
AfUhftip., VoL XY, 1885.
8. SoHWALBs.— Studien Qber Pithecanthropus erectus (Dubois). Zeitachifi far
Morphologie und AfUhrqpologie, Band I, Heft L

4. Der Neanderthalsohftdel. Banner JahrbOcher, Heft 106, 1901.

5. Die Yorgeschichte des Menschen. Braunschweig, Vieweg u. Sohn, 1904.

6. TnBHBB, Sib Williah.— Report on the Human Skeletons— The Crania.

ChaUengcr B^^orts, YoL z. Part xxiz, 1884.

7. Contributions to the Craniology of the People of the Empire of India.

Part I. TransacHons Boyal Soddy qfEdin,, YoL zxxix, Part 8, Na 28, 1899.
Part II. Transactions Boyal Society qfEdin., Yol. zl. Part 1, No. 6, 1901.
Part ni. Transactions Boyd Society o/Edin., YoL zlv. Part 2, No. 10, 1906.

8. A Contribution to the Craniology of the People of Scotland. Part L

Transactions Boyai Society ofEdin.^ 1908.

Additional Note on the Neanderthal Skull. Quartedy Joum. qf Science^

October, 1864, p. 758.

Plate V

Fig. 2

Fig. 1. Schwalbe's method of showing elevation and depression of the frontal bone..
Fig. 2. Mesial contour-tracing of the fore-part of an Australian cranium.

Plate VI

^coMs/f

/y^/3/)

Sco^tis/)

Fig. 4

Fig. 3. Mesial contour-tracing of the cranium of a female Australian.
Fig. 4. Mesial contour-tracings of the fore-parts of three skulls ; showing variations.

Plate VII

o
o

THE PLACE OF THE ' SONDER-GOTTER ' IN
GREEK POLYTHEISM

By L. K. FARNELL, D.Lrrr.

It has been said that all study of popular religion is a study of
popular psychology; and this is true so far as our main object is
to discover the feelings or ideas that underlie the ritual or external
act of worship, the early and often prehistoric thought that inspired
it, as well as the later thought of any given historic period. This
is especially diflScult in regard to a class of cult-figures in Greek
religion that may seem to belong, and have been explained as
belonging, to an older stratum of national belief than that mth
which the Greek student is familiar, a * polydaemonism ' rather
than a pol3rtheism. These figures are in some sense nameless, in
that they seem to have possessed no substantival proper names
but merely appellative epithets which usually reveal the narrow
function or department to which tlieir daemonistic agency may have
been coilfined. As a rule, there is httle l^end attaching to them,
they have rarely a genealogy or family history, but appear as barren
and isolated personalities standing apart from the warm life of
Greek polytheism. They seem at first sight nothing more than
shadowy potencies of the field and fold, of the human household or
state, or somethnes of the arts and higher functions of Ufe, and they
are called indifferently ©eot, Ao-t/iop-c^, ^HpQ)€^. For the purposes of
a general survey, we may classify them according to their depart*
ments. As powers of the field and the crops the record gives us
Ewoa-To^ at Tanagra, the hero who brings a good yield of com,
'ExcrXatos igp^^i *'^® well-known * hero of the ploughshare ' at
Marathon, Kt/a/Lttrij^, the bean-hero, whose shrine was on the sacred
way to Eleusis ; with these we may consider Av^Tjaria of Aegina,
and SaWoi and Kapwd^ the Attic Hours, and *Ept^ota, the cattle*
goddess of Lesbos, and perhaps we may bring into this company the
Zaifimu hnBdTT)^ of Sparta, ' the giver of good gifts/ With the
guardianship of the life of the family and the fostering of children

TIJE PLACE OF THE * SONDER-GOTTER '

are associated certain doubtful personages such as KovpoTp6<f>o^^

* the nurse of children,* KaXXtycVcta, * the giver of fair offspring/
*A/jL(^tSpo/Aos, a Sai'fteai/ whose personality was perhaps invented by
Aeschylus and who arose from the *A/i<^iSpo/xta, a ritual at which the
new-bom child was solemnly carried round the hearth-fire and
named in the presence of the kinsmen ; we may also remember that
Charondas speaks of certain Bai^ove^ cotioS^o*, powers of the sacred
hearth. Sometimes a hero or daimon might protect the gateway of
the house or city or the city- walls or the entrance to the temple, as
we hear of a ijpai^ wpo wvkSnf in Thrace, of an eTrtreyto? rfpm<; and retxo-
<j>vXaS at Athens, the guardian of roof and wall, of K\aiKo<f)6po^, the

* holder of the temple-keys *, at Epidauros. At the banquet, not only
were the high gods remembered, but possibly such personages as
Aatnj^ at the later Ilium, 'AKparoTroTif}^ at Munychia, AetTrvcv? in
Achaia, K€pdmp and Mdrrti^p at Sparta, and if we had only the name
to guide us we might associate mth these the Saifia^p to-oSatnj^, the
daimon who presided ' over the equal feast '. Again, the potter's art at
Athens seems to have required a i7pm9 ^c/oa/io?, the medical a yjpo)^
iarpo's at Marathon, Athens, Eleusis, the nautical a i^poi^ Kara npvfjLpav
at Phaleron, and a yjpio^ crrparTjyos is mentioned in an Athenian
inscription. The enigmatical name BXatJrij occurs on an inscription
of a late period found on the Acropolis, from which we learn that
she shared a shrine with KovpoTp6<f)o^. We might be tempted to
accept KOhler's suggestion that the word is really BXao-rtj, an appella-
tive of a spirit of vegetation. But the letters as they are given in
the Corpus Inseriptionum Atticarum appear to have been correctly
transcribed. If we may trust a gloss in Pollux, there was a jjpo)^
iirl 0XavTg at Athens, and jSXavnj was the name of a kind of sandal,
and we seem to be dealing with the patron saint of shoemakers,
though why such a person should have shared the shrine of
KovpoTp6«f>o^ is not easy to explain. Such figures appear to have
been comparatively numerous in Attica, for to those already men-
tioned must be added the jjpui^ %T€4>av7j^6po^ at Athens, the
l,wovhaimv haifimp on the Acropolis, a kindred personage to the
haifiutv iwtBamFf^ or the *Ayado^ iaifjLOtVj and TcXecriSpo/io^ at Eleusis,
apparently a hero presiding over the athletic contest in the
Eleusinian festival. At Delphi, a parallel figure to TcXccrtSpo^o? has
been discovered in EiUSpo/xo^, whose chapel is attested there by a
fifth-century inscription, the hero to whom the runners prayed. At
Lesbos we recognize a daimon of the weather, whose function possibly

IN GREEK POLYTHEISM

was to give the favourable breeze, in 'Err)ff>iXa or XIj/toTta 'Erry^tXa,
mentioned with Poseidon in a long ritual inscription. And at
Knidos the *Emfiaxo^ mentioned in an inscription already noticed
may belong to the adjacent name of Pluto, though it does not seem
to be an epithet natural to this god, or it may be the appellative of a
distinct cult-figure. The list closes with the names of two whom we
should rather expect to find in the Roman Indigitamenta than in a
catalogue of Greek heroes, the hero * who fi-ightened horses * in the
race-course at Olynipia and on the Isthmus, and the * Fly-catcher * at
Aliphera in Arcadia, Tapd^iTnro^ and Mviaypoq,

To the same stage of religious psychology at which these cult-
figures might seem likely to have developed may have belonged
those vague groups of divine personages that are also characterized
by a functional appellative rather than by a proper or substantival
name ; for if the single functional daimon appears to lack
individuality and concrete personality, compared with the high
gods and goddesses of polytheism, groups of such characters united
only by a single fimctional name will be likely to be still more
shadowy and amorphous. While detailed criticism of these may
be reserved for the present, the following list presents them in
alphabetical order.

The 0€ot *AwoTpQTTaLoi were worshipped at Sikyon near the
grave of Epopeus, the mythic ancestor; and, as Pausanias tells us,
rites were performed to them such as were usual among the Greeks
* for the turning aside of evils ' : his words imply that there were
images of them erected near the grave : the 0eot Fci'ervXXtSe? and
KciuXiaSes were deities of childbirth much worshipped by Attic
women, greatly to the sorrow and cost of the husbands, if we may
trust Lucian : the EiBdp€fioi appear to have been a group of weather-
daimons or wind-charmers, to whom an altar was consecrated in the
Kerameikos and apparently another at Eleusis. The Bcot Ka^apol at
Pallantion in Arcadia are the subject of a very interesting note in
Pausanias : ' there is a temple of 0eoi still standing on the top of the
ridge ; they are called KaOapotj and oaths on matters of the greatest
import are taken before them. The people do not know their names,
or knowing them are unwiUing to pronounce them. One may con-
jecture that they were called Ka^apot' because Pallas offered to them
a different kind of sacrifice from that which his father (King Lykaon)
offered to Zeus Au^ato?.' Pausanias has probably the Dehan altar in
his mind that was called Ka^apd^ because no blood was ever shed

o 2

THE PLACE OP THE * SONDER-GOTTER '

upon it. The Seal MetkCxt^oi at Myonia in Lokris may have been a
similar concept : we can gather that they were chthonian powers, to
whom rites of purification for sin, probably the sin of bloodshedp
were performed by night. Certain Oeol MuXavrcioi are mentioned by
Hesychius and defined as ' deities of the mill ' ; but his explanation is
very doubtful ; he elsewhere speaks of a Tlpofiv\€v<;, a goddess whose
statue was erected in corn mills. More important is the worship of
the Beol IXpaftStVat on Mount Tilphossion, near HaUartos. Pausanias
mentions their hypaethral temple there and adds that the oaths
taken in their name had the most binding force. It may have been
a Minyan migration from this part of Boeotia that brought the cult
to the shores of Laconia near Gythion, where Pausanias found in
the popular tradition the reminiscence of a «^€a UpaitSiKay whose
cult was associated with the return of Menelaos from Troy. The
significance of the name is obvious ; the Ilpaf tStVat are local variants
of the 'Epii/vc9, their appellative expressing more clearly the abstract
conception of moral retribution. The ^ap/jta^tSc? at Thebes may
once have been the vague pei^sonages of an early cult, and akin to
the EtXcWuiat, the divine powers that could aid or retard childbirth.
Before raising any further question about such groups, or consider-
ing how the conception of divinity that attaclies to them diflfers from
that of ordinary polytheism, it may be well to put oneself on one's
guard. A divine group united by some common appellative may
have consisted merely of some well-known high gods, whose
figures were as concrete and well defined witliin the group as without
it For instance, the term Beol ^Ayopaloi certainly describes no
shadowy company of half-formed Sat/Ltoi/c?, but denotes the deities
whose statues happened to stand in the *Ayopa, and these were
usually Zeus, Hermes, Apollo, Athena Or again, the Beol UpoSofxel^,
who were worshipped at a iaria in Megara, may indeed have been
a group of nameless * functional ' Sat/Aoi/€?, who had to be appeased
before the building of cities ; for, according to the legend, sacrifice
was offered to them by Alkathous, the founder of Megara, before
he began to erect the wall : but the context suggests that Apollo,
the god who was pre-eminently the city-buOder, was one of them,
and that we should explain them differently, as the deities whose
statues * stood before the houses ', such as Apollo, Artemis, or Hekate.
Similarly the Beol <I>ptJrptot at Naples, known to us only through
inscriptions of the Roman period, appear to have been worshipped,
not as the heroic ancestors of the clans, but as the deities who

m GREEK POLYTHEISM

presided over the organization of the phratry. In other parts of
Greece these were certain welMefined divinities such as Zeus,
Athena, even Poseidon : and perhaps the group at Naples consisted
merely of such figures as these. Or the designation may have
acquired a certain quasi-Koman vagueness^ and connoted, for
instance, Zeus, Athena and * some others ', the vaguely comprehensive
term being chosen so that no deity might be offended by inadvertent
neglect.

Finally, we can say nothing positive about the *AXitt8ai, the
name of * certain gods in Lacedaemon\ as we learn from the
doubtful authority of Hesychios- If the gloss is correct, we may
have here either the appellative of vaguely conceived divinities, who
never acquired proper names, and were known only as the * mighty
ones ', or the complimentary title of certain ordinary and well-known
personages of Greek polytheism.

It is quite possible, then, that in these latter instances there is
no distinct religious fact that wants explaining. The case may be
otherwise in regard to the other groups : and we must consider these
in connexion with the cults of those separate Sai/xoyc^ or rfpcoe^ above
enumerated, who are known to us only through adjectival appella-
tions, not by any proper or substantival names. The important
question is whether all or some of these are the products of an
earlier prehistoric stage of reUgious thought, a stage of what may be
hypothetically called * polydaemonism *, a conception preceding in
the history of our race the emergence of such articulate and concrete
individualities as are the anthropomorphic figures of Greek polytheism.
Before going further in the examination of tliis question, it is proper
to consider whether the name Sat/x&ii^, which is attached to many of
these indeterminate figures, affords us any clue. The etymology of
this word, even if it were certain, is no sure guide. Its Uterary and
popular usage may be shortly stated thus : in the Homeric poems it
is synonymous sometimes with Bco^ and designates a personal deity :
frequently it expresses for Homer the more abstract divine force,
especially fate or the destiny of man's Hfe, and, in a narrower sense,
the doom of death, Hesiod twice employs it in this sense, and
twice applies it to individual men or demigods who have become
glorified after their death or during their life ; he nowhere clearly
uses it as a synonym for the personal higher gods. In a fragment
of Alcman (69) it occurs in an impersonal sense, meaning apparently
the distribution of human lots, Empedocles uses the term— not

THE PLACE OF THE ^ SONDER-GOTTER '

indeed as an equivalent for the ordinary human soul, as Rohde
supposes— but for the immortal prenatal soul which, having
offended some divine law, is cast out from heaven, and, descending
into a man, passes through a long cycle of existences : with this view
we may connect the later application of it, which is sometimes found,
for instance, in Pindar and Menander, to a man's personal genius.
On the other hand, from the fifth century downwards, it bears two
senses, both of which are concrete and anthropomorphic ; the
Tragedians can designate as Saiftoji' the deceased hero or heroine,
Darius or Alkestis ; and the popular usage was often in accord with
them, for the ferocious spirit of Temesa was a Salfioj^^ but he was
also the *H/)aj9, the companion of Odysseus who was slain by the
inhabitants. Finally, the word came often to denote an inferior or
subordinate deity, as in a Dodonaean inscription we find ^'col rjpo)€<;
Saifxove^ given as a full classification of all the divine powers to
whom prayer or sacrifice might be offered* In this sense Attis and
the Korybantes are called AaifjLOP€s.

It may be that Seoi and dat/Aoi/c^ have both been handed down
from an equally ancient stage of Hellenic speech, both applicable in
the same sense to * gods * : and we may find instances in other
languages for the co-existence of synonyms expressing the same idea
of divinity. The terms will probably tend to differentiation, as, in
fact, ^aifiitiv became variously differentiated.

But the origin of terms does not concern us here. It is sufficient
to note that when applied to these cults which we are examining —
of which the record is comparatively late^it need not be regarded
as investing the cult-figure with a vaguer or more impalpable or
abstract character than that of the Olympians themselves. On
the other hand, we must lay streas on the fact that most of these
personages in the scanty list given above, which I have endeavoured
to make complete, are designated as r}pa}€q : and the value of this
term for the popular imagination is at least clear : it denoted a
glorified man once existing upon the earth. Therefore the *Hpc«J9
*EffxTcyto? or ^Ej^crXatos is qua 'Hpcus as real and palpable a
personage as Apollo or Hermes. If his personahty is to be re-
garded as a survival from a period of vaguer and more amorphous
reHgious conception, it must be on the ground of his designation by
a mere appellative and the absence of a pei'sonal and concrete name.

The facts so far set forth have been made part of the founda-
tion of a far-reaching theory promulgated by Dr. Usener in a

IN GREEK POLYTHEISM

treatise on the Qotternamen, a work of importance and value, of
which the main results appear to have been rather widely accepted.
This is no place for detailed criticism, but some consideration of his
leading principles and conclusions is essential here. He correlates
the Greek facts with the Roman Indigitamenta and certain
phenomena he has observed in the Lithuanian religion ; and the
conclusion towards which he draws is that the Indo-Germanic
nations, on the way to the higher polytheism, passed through an
earlier stage when the objects of cult were beings whom he design
nates by the newly-coined words * Augenblick-Gotter ' and * Sender-
GOtter*; that out of these the * Olympian order*, the concrete anthropo*
morphic gods of Greece and Italy, of the Indo-Iranians, the Per-
sians and Slavs, were evolved, whose more vigorous personalities
absorbed the earlier and vaguer forms, and whose concrete proper
names now attracted to themselves the mass of adjectives and
epithets that were once the independent and sole designation of the
older divine beings ; finally that traces of this evolution can be found
in certain later survivals of the historic cults.

Now the importance of the theory very much depends on what
we mean by a ' Sonder-Gk>tt *. Dr, Usener develops his definition
from Varro's phrase — certi dei — which occurs in a passage of
Servius ^ : ' pontifices dicunt singulis actibus proprios deos praeesse,
hos Varro certos deos appeUat,* He finds the essential characteristic
of a Sonder-Gott, first, in the narrow limitation of his nature or
concept, which seems relative only to a particular act or state, or
even to a particular moment in that act or state ; secondly in the
open transparency of the name which, whether substantival or
adjectival, expresses just the single function that the divine being
exists to perform.

So far we may accept this as prima facie a fair account of the
complex Roman system which is presented in the Indigitamenta.
We owe the statement of this system to the Cliristian Fathers,
Arnobius and Augustine, who reproduce Varro, and Varro appears to
have drawn from the pontifical books. As regards the absolute
authenticity of this record, I cannot express an opinion : it may be
that some of these appellatives in the Indigitamenta are only thin
disguises of well-known concrete gods, such as Faunus and Jupiter,
as an American scholar, J. B. Carter, has endeavoured to prove
in a treatise *de deorum Romanorum cognominibus *. But, if we

^ Aen. ii, 14 L

THE PLACE OF THE ' SONDERGOTTER '

accept the main accoimt of Varro as authentic, we may well sympa-
thize with St, Augustine's humorous protest against the abnormal
* religiosity ' of the Romans that seemed to leave nothing to unaided
human initiative. And it is very difficult to find the right expression
by which to designate this system in terms of the ordinary nomen-
clature of anthropology. It cannot be called fetichism, still less
pantheism. If it really was to the Roman as it appears to us, we
may be tempted to regard it as a very abstract and spiritual form of
animism. If it be a right account of animism that it endows
inanimate and material objects with quasi-human consciousness and
emotions, and sometimes with a supra-human power and voUtion
which suggests womhip^ we may perhaps extend the term to cover a
religious system that imagines an immanent semi-conscious or sub-
conscious divine potency to reside in passing acts and states of man
or fleeting operations of nature.

This leads us to the next consideration, which is of still greater
importance. Are these * Sonder-G5tter ' conceived as personal gods ?
Dr. Usener does not always speak quite clearly on this point ; he
maintains, on the one hand, that a few of them can be proved to
have had a personal reality for the ItaUans, yet his tendency is to
distinguish this Roman system, which he finds also in Greece and
Lithuania, from the polytheistic belief in personal gods. If this
distinction on the ground of personality is justified, it is vital ;
because in tracing the evolution of rehgion, and in classifying
recorded or existing forms, the most far-reaching principle of
classification is the distinction between the anthropomorphic and
non-anthropomorphic fornis of beUef, the personal and the imper-
sonal or half-personal objects of reverence.

Supposing, then, that the above-given account of the Sonder-
GStter is correct, have we the right to regard them as belonging always
and everywhere to that more primitive stage of belief which preceded
polytheism and led up to it ? Looking first at the minute specializa-
tion of divine functions on which the system is based, we cannot
regard this as a decisive test of primitiveness. Such specialization
may indeed be found among early races, nor am I inclined to believe
in the neo-totemistic dogma *one clan one totem-god'. Some of
Dr. Usener's Lithuanian parallels may be accurate illustrations of the
species that he is formulating^ though I do not recognize the value of
all of them ; certainly * the Fly-Buzzer God \ a Lithuanian form of
Mvlaypo^^ the * God of the Besom ', the ' God that makes the grass

IN GREEK POLYTHEISM

green ', the ' God who makes the beer sour \ these are deities with a
distinctly Roman flavour about them. Having tried to go further
afield I have been able to find only a few exact parallels* Dorsey,
in his ' Study of Sioujt cults ', mentions the Indian s invocation of his
hunting-trap and all the various parts of it, and his prayers to the
tent-pole, which are quite after the fashion and spirit of the Roman
Indigitanienta. Traces of the same system seem to appear in the
religion of the Kenyahs, a tribe on the Baram river in Borneo,
described by Messrs. Hose and McDougall ^ : * Balli Atap (Atap = roof)
is the spirit or god that protects the household fmm harm of all sorts,*
and reminds us of the "Hpcn? 'ETrWeyto? at Athens ; and in the prayers
of certain heathen tribes in Russia we may detect the same * Indigita-
menta' style, ^^ But I imagine we should find this rigorous appor*
tionment of special functionB, this minute articulation of the divine
world, at least aa frequently in the latter days of a well-organized
polytheism, of which it is often a mere by-product. While many
of the personal gods in Greece expanded their individualities and
widened their range of functions, many were obliged to contract and
to speciali^. Ares and Pan were once more manifold gods than
they afterwards became ; and the same is true of Aplu*odite and
Eros, and in some degree of Artemis. And such personal deities as
Eros and Asklepios beget such transparent and limited personages
as Himeros and Pothos, laso Akesis, Panakeia : while Nt«7j, Ilct^ci,
NcfLco-c-?, most absolute Sonder-Gdtter, are late products of poly-
theism, and the first two, if not the third also, are probably emana-
tions of concrete and personal deities.

The specialization of functions, then, is not a test that helps us
to distinguish the ' Sonder-GOtter ' system from personal poljiiheism,
or to assign the former of necessity to a more primitive stage. But
the greater or less degree of anthropomorphism in these strange
Greek, Roman, and Lithuanian forms, if we could appreciate it,
would be a much more important clue. And it is in dealing with
this question that Dr. Usener's work appears least satisfactory.

It is obvious, as Mr. Warde Fowler and other writers on Roman
religion have often pointed out, that it was fai* less anthropomoi-phic
than the Greek, that it presented leas concrete individualities to the
imagination. The chief deities of the Italic tribes were personal and
anthropomorphic in so far as they were distinct in sex and were

' Joum, Anthrop. Inst, 1901, pp. 174-5.

' Arthiv fur Beligumsmssenschqftf 1906^ p, 284.

THE PLACE OP THE ' SONDER-GOTTER *

worshipped occasionally with idols ; but the high powers of the
Koman rehgion seem to stand apart, each for himself or herself^ in
pr a cold aloofness. Little or no myth is told of them, rarely a legend
of marriage or affiliation. Were, then, Inuus, Occator, Dea Panda,
Deus Lactans, Dea Mena, and all the crowd of deities of procreation,
nutrition, and birth, invested with a personality very much vaguer
and thinner than were Vesta and Minerva ? And, if so, are they to
be regarded as the survivals of an older stratum of religion, or rather
as the late development of a certain logical tendency in Roman re-
ligious thought ? The record is late, and gives us little more than a
bare list of names ; and no clue is offered by any tradition or any
reported ritual. Nor is this a place to attempt the solution of the
Roman problem.

As regards the Lithuanian evidence, the exposition of it by
Dr. Usener fails to show the different degrees of strength with which
the various functional agencies in his list were personified, or to
distinguish between the more concrete and the vaguer forms. It is
very interesting in itself, but I do not think it solves this particular
problem of Greek polytheism.

We can now confine our attention exclusively to the Greek
evidence. We have every reason to beUeve that the Hellenic per-
ception of divinity had become concrete and precise at a very early
period ^ ; even if theriomorphism occasionally prevailed, the clear
outlines of the divine personality need not have been much impaired;
there is nothing necessarily vague or nebulous about a horse-
headed Demeter. Moreover, the chief divine personalities had at an
early period become anthropomorphic. The view is quite tenable
that many of the anthropomorphic deities were already the common
possession of the Greek tribes before the migration into Hellas*
The extreme antiquity and obscurity of most of their personal
names would itself support this view. And the impulse in Greek
religion towards the creation of clearly outhned personal forms was
a devoming impulse that might well have obliterated the traces of a
previous more amorphous animistic system. Yet such traces may
be found, and m other directions more clearly perhaps than in the
domain of the * Sonder-GOtter '. The worsliip of the stone, the
pillar, the tree-trunk, even the axe, is proved of the prehistoric
period, and it survived in the historic. It is sufficient to observe
here that such aniconic cults are compatible and often contem-
^ Dr. Uaener himself adxujts this, p. 302.

IN GREEK POLYTHEISM

poraneous with an anthropomorphic and personal conception of
the divinity, though they may have arisen under the influence
of animism, fetichism, or from mere ' teratology 'J Thus the
* Mycenaeans ' possessed human and personal gods, though their
iyaX/jtara were the pillar, the tree, or the axe : as witness we have
the sacrificial scene on a Mycenaean gem, possessed and recently
published by Dr. Arthur Evans, where a god is seen hovering
above his own pillar, having been evoked by the prayers or the
ritual. But the Arcadian cults of Zci? Ktpawo^j Zeig KamrciTa^, in
wliich Zeus was actually identified with the thunder and the meteor-
stone, and the fetich-worship of the sceptre of Agamemnon at
Chaironeia, seem to belong to some primitive stratum of pre-
anthroponiorphic religion. We must beheve in the existence
of this stratum in the buried soil of the Hellenic or pre-Hellenic
religions as a 'vera causa' that might explain certain anomalies
among the religious facts of the historic period.

But it is very doubtful if we need invoke the aid of this hypo-
thesis to explain the facts upon which Dr, Usener has built his
theory ; and there are some that it would fail altogether to ex-
plain. There is one important point that we must insist on at
the outset, A god is not necessarily nameless because he is not
named or is usually addressed by a simple appellative. There are
many reasons for concealing the proper name. One is the super-
stitious fear that the enemy may come to possess it, and work evil
through the magical power that the possession may give him. For
the same reason many savages conceal their own true name and the
names of their friends ; and this is occasionally found even in
civilized communities ; as, for instance, it was improper to mention
the personal name of the S^^SoS^o? ^t Athens on account of his
sacred character. Again, it was illomened to use the name of the
deities of the nether world, because of their associations with death.
Thus arose euphemisms for the name of Hades ; and the designation

* Statements about the animistic worship of stones and trees are often
deoeptive ; the words of Miss Alice Fletcher in the Peabodtf Mmeum Reports,
vol. iii, p. 276, * Careful inquiry fails to sho^'r that the Indian actually worships
the objects that are set up or mentioned by him in hia ceremonies. Tlie earth, the
four winds, the aun, moon and stars, the stones, the water^ the various atiimals^ are
all exponents of a mysterious life and power encompassing the Indian and filling
him with vague apprehension and desire to propitiate. . , . These various objects
are stopping-placea of the god/ may serve as a correction of hastOy gathered
impressions*

THE PLACE OF THE * SONDERGOTTER *

of the god and goddess of the lower world as o Seo^ and 17 ©ca, which
came into vogue at Eleusis in the fifth century BvC.^ may be due to
the same motive, and need not be supposed to have descended from a
system of nameless deities of dateless antiquity. A similar feeling
prompted the habit of passing the graves of the dead, and especially
of the dead hero, in silence ; and from this practice the buried hero
at Oropus received the name StyijXd?. And as many heroes came
thus to be designated simply as 6 ''Hpm% the personal names could
easily pass out of recollection. What was superstition in one age
becomes merely respectful reserve in another ; and the modern man
rarely speaks of God by any personal name, but most frequently by
some vaguer title such as ^ the Deity \ At Bulis, near Phokis, the
chief god was always addressed merely by the worshipftil title of
Meyia-TQ^j and never by any proper name, according to Pausanias ^ :
but there is no reason to suppose that they had not advanced as
far in the evolution of anthropomorphic and concrete divinities as
their neighbours, or to gainsay the view of Pausanias, that Mcytcrro?
was none other than Zeus himself.^

We may next observe that many of the divine appellatives that
Dr. Usener presses into the support of his theory are no signs of any
earlier and distinct reHgious stage at all, but are as anthropomorphic
in their connotation as any individual proper name, and many have
a generally descriptive and no functional sense whatever, and there-
fore are by no means to be compared with the Roman Indigitamenta.
For instance, we find in him the strange suggestion (which is almost
a reductio ad absurdum of his theory) that Demeter BapBiq derived
her appellative from an old god called Hai^^o^ ; the only person so
named was a secular hero, and there is no evidence of a divine
personage so called except for those who hold, like Dr. Usener, the
almost obsolete and very narrow theory that all popular heroes of epic
and legend were the faded forms of forgotten gods. But let us grant
a god Hai'^o?, or a goddess SavBij. There is nothing * functional ' about

' 10, 87, a

* It is partioularly in the Eastern Heilenised world, in various districts of
Asia Minor y especially Phrygi% that we mark the tendency gaining force in the
later period to designate the divinity by a vague descriptive name of reverential
import, such as * the Highest God ' ; two newly discovered inscriptions of the
Eoman period at Miletus show the existence there of a cult of aytwraro^ $€0^
^\f/ifrro^ ^t»m}p, who was a god of divination and sei-ved by a vpo^ijnys. — Arch. Ans,^
1904, p, 9.

IN GREEK POLYTHEISM

the adjective name, nothing vague : it has more obvious anthropo-
morphic connotation than the names Apollo, Athena, &c. It no
more marks a distinct stage in religious thought than two such
formally different names of individual men as * White ' and ^ Wright '
mark two different stages in the development of our personal con-
sciousness concerning our fellows.

Still less relevant to the hypothesis of * Sonder-G5tter ', or a
system of specialized functional divinities vaguely and almost im-
personally conceived, are such popular titles of divinities as Sctirctpa,
Accnroti/a, BacriXij, Was there ever an imaginable stage in Aryan
religion when deities were brought forth immaturely with noth-
ing more concrete to cover them than the vague * function/ of

* Ladyship ', ' Queenship ', * Saviour Power ' ? Surely such names
are the natural adjuncts of personal religion, and belong to the
ceremoniousness of personal worship. Salrctpa is here Kore, there
Artemis, elsewhere Athena ; it is certainly difficult to imagine her
before she was any one at all in particular. And if we could, we
still could not call her a Sonder-Gtittin according to the definition.
In many parts of the Mediterranean, long before Christianity, a
virgin-goddess UapB^vo^ was worshipped and known by no other
name. Yet she need not have been evolved to fulfil no other

* function ' than to be maidenly, but probably had in the people's
imagination as marked an individuaUty and as concrete a character
as the Holy Virgin in our own reUgion. We should scarcely say
that the proper name * Mary * and the appellative * Holy Virgin * re-
veal two distinct stages of religious thought. The Goddess 'Aptonj,

* the Best/ may have been worshipped at Athens, Metapontum, and
Tanagra, \vithout a prdper name, but may have been as personal an
individual as Artemis,

In fact, apaxt from the above considerations, the number of
deities and heroes in Greece who can be proved to have existed in
cult without a proper name is exceedingly small. Dr. Usener en-
deavours to enlarge the stock by what appears to me to be faulty
logic ; by the suggested rule, for example, that when two or more
deities have the same epithet in common we should conclude that
the epithet had a separate previous existence as the appellative of a

* Sonder^Gott \ The cogency of tliis does not appear ; every personal
deity was liable to be called *AXc^tVaK09, every goddess or heroine
AtTrapa/ATrvf or BaffvKokwo^. More than one Greek divinity was
called MeiXtx*^o9, a term usually connoting the character of the

THE PLACE OF THE * SONDER-GOTTER *

nether-god, and we have a cult-record of o MctXtx*'*^^? ^ ^^ have of
6 0€O9 alone. But thLs is no reason for supposing that Zeus MctXtxto?
became so by absorbing an older and vaguer 'numen' called ' MctXtxto?'
who had once half-existed in shadowy independence ; for we note
that MetXt;^io9 is a word of later formation within the same language
than * Zeus \

Again, his theory does not sufficiently appreciate the important
factj of which, however, he is cognisant, that we can already discern
the bright personal deities of Greek polytheism throwing off their
epithets as suns may throw off satellites, the epithets then becom-
ing the descriptive names of subordinate divinities or heroines.
Examples of this process have often been given and discussed. It
is a tenable belief that Aphrodite threw off Peitho, Athena Nike,
Poseidon Aigeus ; the most transparent fraud of all was the emanation
of a useless and colourless hero JlvBio^ from Apollo Ilu^to?. In
Thera the people were especially prone to call the high gods by
their appropriate appellatives. The inscriptions* show an Apollo
A€X<^iVi09 styled AcX^ti^io?, Zeus 'l^cVto? Srotj^ato? IloXtcu? *Op/cto?
invoked by these epithets alone. The nether-world god becomes
addressed as *the Rich One*, 'UXovTmp\ ^ He of good counsel,'
Euj9ovXcu9, the * Placable One ', * MetXij(£09 '- Adjectives are more
affectionate and the people love them ; they are also a shorter style.
The process of detaching an epithet from a deity and forming from
it a new divine personality is found also in the Vedic religion.
* Rohita, originally an epithet of the sun, figures in the A V, as a
separate deity in the capacity of a creator/ ^

Bearing these facts in mind, we may now consider again in
detail the short Ust of * functional * and appellative heroes, daimones,
or gods, which was given at the beginning of this paper. We shall
rarely find that they accord with the definition of Sonder-GOtter
or betray a pre-anthropomorphic imagination. The heroes of
the drinking-bout and festive meal, 'AKparowoTT)^^ AaitTj^ Aetiri^ew,
KcpacD^, and Mdrrmv^ are functional, but being heroes are con-
ceived as personal and human ; and none can be said to savour
of prehistoric antiquity^ but are obviously late creations. As there
was no high god that had charge of the banquet, Greek polytheism,
following its natural instinct, creates Aatnj? and Auwpev^^ and
obeying its overpowering bias towards anthropomorphism and con*

" CLG.f Itis, Mar. Acg,, iii, p. 80. ' Macdonell, Vedk Ritual, p. 115.

IN GREEK POLYTHEISM

Crete forms conceives of them as heroes ; and as it was necessary
to invent a name it was more natural to choose appellative descrip-
tive names than to coin in-elevant proper names. Nor is it incon-
ceivable that * AKparoTTorr)^ was a distant descendant of Dionysos
'AKpaT0ff>6po%^ who was known at Phigaleia As regards Kepdcav and
Marraii/j I venture this explanation : the guild of cooks, like other
guilds and like clans of kinsmen, would be tempted to invent for
themselves an eponymous ancestor ; so fictitious heroes arise, whose
names stamp them as the patron-saints of the arts of cooking. We
can similarly explain Kcpa/LLo^ as the eponymous hero of the potters'
guild, who gave his name to a deme of the Akamantid tribe. Nor
must we take these fictions too seriously.

'A/if^tSpo^os we may regard as a pure literary invention, created
to explain the *A/i,<^tSpc5/Ata, as "Eporrf has been supposed to have been
evolved to explain the *Ep<7ij<^pta. The Sai/ioii/ cVtSa>r»j9 of Sparta,
a vague figure with a semi-functional name, certainly seems to
answer somewhat to the description of a true Sonder-Gott ; but the
record of Pausanias suggests that his title is of late creation. The
BaifjLwp XnovSatcjp on the Acropolis at Athens may be regarded as
another form of the 'Aya^o^ Aatfttui^, a late growth of the polytheistic
period. As regards such personages as BXaiJnj, ^Upat^ 'ETrtreyto?, we
have no clue at all as to their character, period, or raison d'etre. More
interesting are the figures of Evvotrro^ at Tanagra and 'E;(CTXato9 at
Marathon, popular local heroes of the field and crops, to whom cei'tain
vivid legends are attached that place them on a different plane from
the shadowy figures of tlie Indigitamenta. The Marathonian tradi-
tion is well known ; it is probably a pseudo-historic aetiological story
invented to explain a name and a half-forgotten cult, and should not
be regarded as proof that the latter originated in the fifth century b.c.
We have still more reason to believe that the Tanagran Eunostos
belonged to a very early period of European behef, and the study of
his legend and the names associated with it reveals an old-world
agricultural story and ritual. Eunostos is the power that gives
* a good return * to the crops ^ ; and, if we may trust the Etymo-
logicum Magnum, he had a sister Evvo(rTo<;, a mill-goddess, who
looked after the measure of the barley, and whose image stood in the
mills, Plutarch tells us that the holy grove of the Tanagran hero
was strictly guarded against the intrusion of women. We know this
to have been a taboo enforced in many ancient shrines ; but Plutarch,
' Cf the use of vo<rro^ in Athenae — 618 C.

THE PLACE OF THE * SONDER-GOTTER '

drawing from a book by Diocles ircpl tZv Jipmmv and ultimatelj from
the Boeotian poetess Myrtis, gives a curious story to explain the fact
A maiden of the country woos the \drtuous Eunostos in vain, and
thereupon hangs herself in grief. To requite her death one of her
brothers slays Eunostos, whose ghost then becomes a scourge to the
territory until he is pacified with cult and a shrine where no women
might enter. The rule was once infringed, with the result of earth-
quakes, famine, and other prodigies, and Eunostos was seen hastening
to the sea to cleanse himself from the pollution. This genial tale of
despised love doubtless arose out of a quaint agricultural or horti-
cultural ritual/ Eunostos is the hero of the cornfield, who is slain
like John Barleycorn is slain. His parents are 'EXtcus of the marshes
and Sicuis of the shade ; the wicked brother is BovicoXo^ ; the hapless
maiden is ^0\va^ the ' Pear-tree ', and these hanging-stories of person-
agee, whose names or legends convey an allusion to the fertility of
the trees and the crops, arose, as I have pointed out before, from the
old agrarian ritual of hanging images on trees. We may then regard
Eunostos and Echetlaios, possibly also *E^ex^€V9, *the ground-
breaker', as descendants or survivals of a very old stratum of Euro-
pean agricultural religion, when the personages of worship were
simpler in their structure and less individualized than the high gods
of Greece ; yet as we know them these Greek heroes of the field and
the tiee are of the same concrete life as that which quickened the
forms of Hermes and Dionysos, Going back as far as we can, we have
not yet found among them the shadowy impalpable forms that seem to
float before us in the Indigitamenta* Kva^tnjg, the bean-hero, whose
shrine stood on the sacred way, may have had the same descent and
character as Evvootot; or he may be a late product, a personage
who grew up artificially within the area of the Demeter-cult, at a
time when the passion for hen>worship had reached the pitch that
it had attained in the seventh and sixth centuries, and culture-heroes
were needed for many departments of life ; he may also have been
called into existence because the culture of beans could not be
imputed to Demeter, who happened to loathe them. Telesidromos,
the hero of the Eleusinian racecourse, is obviously a late and trans-
parent fiction^ and we may believe the same of EuSpofios of Delphi.
Agaui^ we must reckon with the possibility that the theory of
EuhemeroB may occasionally have been true- The worship of real
people of fle^h and blood is a living force, as Sir Alfred Lyall has
emphatically pointed out, in India and China to this day. He

IN GREEK POLYTHEISM

records the case of the very real Indian, Hurdeo Lala, becoming
after his death the * functional ' god of cholera.* It would be quite
natural, from the Greek point of view, that when an individual was
deified or * heroized ' after his death a new and functional name
should be then attached to him, expressive of the benign influence
which he was called upon to exert in behalf of liis woi^hippers.
This would explain such family cults as those of Epimachos at
Knidos and Erythrai, and of Symmachos at Pharsalos. That these
are the cults of real men is certain in the latter ^ case and probable
in the former. Similarly, the cult of the "H/^cu? Sr/aan^yo^ at Athens,
of which we have proof in the first century b. c, may well have been
the cult of a real historical personage whose name was concealed and
lost The *Hpai€9 'larpot in different parts of Attica may with per*
fectly good reason be supposed to have been real men, who had an
existence apart from their * function ', or at least ancestors imagined
and worshipped as real, who take over the art of healing, as every
*hero' always could if he Avished, And of two of these glorified
*laTpoL personal names are actually recorded. The j}pa}q xara irpyfivav
at Phaleron need not originally have been the functional demon-
impersonation of steering, but a buried and sacred personage whose
name was lost, and who was believed to have been the steersman of
Theseus, and thus came to be an occasional patron-saint of mariners.
Greece was full of forgotten graves belonging to an immemorial past.
Many were believed to be, and very Ukely were, the i*esting-places
of ancestral cliiefs, and cults consecrated to them may often have
arisen or been revived after the name had been forgotten. We
know that tombs were frequently near or within the precincts of
temples, and from this local accident the buried ancestor might
acquire a new descriptive name, such as KXal'/coc^opos, the ^ porter of
the temple \ A clear instance of an apparently functional cult
which may be thus explained, and to which Dr. Usener s theory can
be proved inappropriate, was that of the hero Tapd^imro^ at Olympia
and on the Isthmos, Near the entrance to the racecoui-se at Olympia
was an altar wliich appears to have been erected over a grave, where

* Asiatic Studies^ 2nd ser., p» 287*
' B, C. H.y 12, p. 184, On a relief found at Pharsalos, of tbe fourth century
B. a, SymmachoB h seen istanding by his horse, and his type is common for that of
the ' heroized ' dead ; near him b a seated goddess, whom the remains of letters
prove to be Hestia* As the writer of the article points out, this is a unicjue
instance of the figure of Hestia being used as the divine symbol of a family cult.
TTtfOa H

THE PLACE OF THE ' SONDERGOTTER '

we may suppose that some one had l^een really buried, and at this
place horses habitually shied. What was more natural than to
account for their fear by supposing the ghost to be the cause of it ?
It was most important, then, to know the name of that ghost, but
though various theories as to his personal name were put forward,
none could prevail, and the most reasonable course was adopted of
calling him TapdiiTnro^. As the institution of the races at Olympia
is comparatively late, Tapdimiros at least is not a remnant of a pre-
historic rehgion.

Again, there are other appellatives in this Ust that we may
quite reasonably explain as the sheddings and leavings of concrete
high divinities, *EptjSota, for instance, in Lesbos, being very probably
an epithet of Demeter, Ev^otTta or Eiwoo-ia in Phrygia of Agrip-
pina-Demeter. And what are we to say of 'IcroSatrrj^? We can
unden5tand the creation or evolution of a *daimon of the banquet', but
the *daimon of the equal banquet' seems a somewhat stranger fiction.
The record in Harpokration gives us a clue to a different explanation.
We are told that he was a itulKo^ Aat/xtx>i/ at Athens, who was wor-
shipped by women of doubtfiil character. Now, there were certain
foreign cults of Dionysos, mystic and disreputable, that were in
vogue at Athens from the fifth century onwards, and were specially
attractive to women, and Plutarch tells us that in mystic circles
Dionysos was called *IcroSaiT7?.

More important are the cults of KaXXtycVeta and Kovporpo^^o?, both
of whom are ' Sonder-Gottheiten ' in Dr, Usener s Ust, KaXXiycVeta
may be interpreted as the goddess of fair offspring, or as she who gives
fair offspring. The ancient writers, both learned and popular, were
doubtful about her, but all associated her with Demeter ; and it is in
the company of this goddess that we meet with her both at Athens
and in Sicily. The name must be considered in close relation to the
ritual of the KaXXiytVetiz, which took place in the great festival of
Demeter, the Thesmophoria, on the day after the NijcrTcta. Wherever
the Thesmophoria was held in Greece, the KaXXiycVcta must have
usually formed part of it, for Plutarch specially notes its non-
existence at Eretria. It is a legitimate conjecture that on this day
the goddess was believed to have been reunited with Kore, and that
the women then pmyed for fair offspring to the goddess of fair
offspring, the SecrfLotfyopia being specially a festival of married women.
But Kalligeneia herself was almost certainly a later fiction like
Amphidromos, an iniaginaiy personality invented to explain the

IN GREEK POLYTHEISM

name of the festival-day, ra KaXXiyeVclaJ As regards Kovporpoif^o^
at Athens, we cannot be sure that she did not once possess a more
concrete proper name. Many goddesses were called by tliis adjective,
and the type of a female divinity holding a child in her arms, or
giving suck to it, was widely spread over the Mediterranean at a very
early time, and it has recently been discovered that Aphaia in
Aegina was thus represented. The very multiplicity of the proper
names that might claim the epithet might be a reason for a cautious
cult preferring to use the epithet alone. But in any case the
KQvpoTp6<f>os at Athens was a robust and personal figure closely akin
to the earth goddess, and whether the earth-mother is called Ge or
Kovporp6^o<;^ or XlavBmpaj the conception may be equally anthropo-
morphic and personal in each case, and this is really the important
fact to bear in view,

I have reserved for the close of this short critical account
the consideration of Mviaypo^, the Fly-Catcher, at Ahphera; for
Dr. Usencr's theory might really make more out of this humble
personage than out of any of his confreres. The facts that illustrate
the cult are interesting. At Leukas and Actium they sacrificed to
the flies before tlie great ritual in honour of Apollo began. This
was perhaps the simplest and most primitive thing to do ; it does not
imply fly-worship, but the preliminary oifering to them of a piece of
cooked meat was a bribe to the flies to go away and not disturb the
worshippers at the solemn function that was to follow, where any
disturbance would be ill-omened, and where the flies were likely,
unless pacified first, to be attracted by the savour of the burnt-
sacrifice. As thought advances, a hero, Mvtaypo^, is evolved at
Aliphera, to look after the fiies before the sacrifice to Athena We
have traces of the same hero at Olympia, though here his function
was at last absorbed by Zeus 'Awo^vio^. Here then in Mviaypoq is
almost the trae Sonder-Gott, almost the * Augenblick-Gott * ; for his
function is very limited, and his value for the worshipper was prob-
ably little more than momentary, nor are any stories told about him*
Yet he is a late invention, implying the pre-existence of the higher
gods, for whose better ministration he was created and ordained.
Likewise he is called a '^p<i}<; by Pausanias, and therefore by him at
least regarded as personal.

It seems, then, that scaicely any figure in this brief catalogue
entirely satisfies Dr, Usener s definition of a Sonder-Gott; those that
^ Vide my CultSy voL iii, pp. 95-6.
H 2

100 THE 'SONDER-GOTTER' IN GREEK POLYTHEISM

may be supposed to have descended from a remote past yet possess
a personal character which betrays the same religious thought as
tliat which produced the pei-sonal gods of poljiiheism. Many of the
figures imply the high gods, and some are probably emanations from
them. The more shadowy and impalpable forms can be sometimes
proved, and often suspected, to be the products of the latest period.
The ancestor of a personal deity may be often more limited in
fimction, but appears sometimes to be more complex than iiis
descendant. An adjectival name may have been originally chosen
to designate the Godhead; the name 'Christ* was adjectival, and
originally ' Zeus ' may have only signified * the Bright One \ But
snch names may in thought have been connected with many other
qualities that make up personality, and may have at once denoted
full concrete individuals* Doubtless a divine individuahty often
grows in the coui-se of time more complex and more intensely con-
ceived, and sometimes we can mark the stages of its growth. But
Dr. Usoner's learned and, in many respects, valuable treatise has
not proved, or even made probable, its theorem that in the immediate
background of Greek pol}i:lieism, out of which much of it developed,
was a shadowy world of functional, half-impersonal * numina \ Greek
religion early and late had always its animistic and daemonistic
elements ; and in the history of our race animism probably preceded
theism and polytheism ; but our present knowledge points to the
beUef that the ancestors of the historic Greeks brought with them
a personal religion of concrete divinities, and found a pei-sonal poly-
theism in many respects differing from their own, but in other ways
akin, on the soil that they conquered.

FOLK-LORE IN THE OLD TESTAMENT

By J, G. FRAZER

It is now generally admitted that the ancient Hebrews did not
attain to the high-water mark of their reUgion and morality at a
single bound. Like every other people they passed through a long
period of development before they rea<:'lied those lofty conceptions
of the divine nature and its relation to man which are the glory of
Israel. The rising tide, if I may pursue the metaphor, did not flow
onward with one broad unwavering sweep ; it had many backward
eddies^ many of those retrograde movements which in the language
of the Bible are familiar to us as backslidings. So the great rollers
break in thunder on a pebbly beach and then retii'e with a griding
sound of pebbles which the retreating water sucks back with it
into the sea. At such times we often doubt whether the tide is
flowing or ebbing. So it must often have been with those who
lived through some of the great epochs in the history of Israel
They also must have had many misgivings as to whether the
movement of thought and conduct was on the whole forward or
backward, whether the changes they witnessed would in the end
prove for good or evil The writings of the Hebrew prophets are
full of these doubts and anxieties. They reflect a state of mind that
seems to tremble on a knife-edge, to oscillate between hope and
despair. From the brightest visions of future glory and bliss we
plunge suddenly into the gloomiest forecasts of coming disaster and
woe. It would be a great injustice to the prophets to imagine that
these dark forebodmgs were nothing but the gigantic shadows of
Assyria, Babylon, and Egypt cast athwart the little land of Palestine,
nothing but fearful anticipations of lost battles and national iiiin,
Tlie prophets were patriots certainly, but they were much more.
They were ethical teachers who viewed with burning indignation
the base and cruel superstitions to which many, if not most, of
their countrymen were slaves. To the best minds of Israel that
moral bondage was worse than any merely political servitude could
ever have been. So they never wearied of denouncing it in language

102

FOLK-LORE IN THE OLD TESTAMENT

which by its fervour and insistence aflfords ns some measure of the
depth and extent of the evil that drew forth such fierce invectives.

In point of fact we learn from the prophetic writings as well as
from the historical books that a mass of paganism ^ and a very gross
paganism too, survived in Israel down practically to the close of the
monarchy. The last great reformation of Jewish religion took place
under King Josiah less than forty years before the capture of Jerusalem
and the final destruction of the national independence ; and down to
that time, as we know from the Book of Kings, some of the worst rites
of heathendom were practised at Jerusalem and even tolerated in the
temple itself.^ Thus we have evidence, abundant evidence, in the
Old Testament that lieathen superstitions persisted among the Jews
to a late era* Such relics of barbarism we are accustomed to call
survivals^ because they have survived from rude ages into a period
of higher culture despite all the humanizing and enlightening
influences that have been at work. It is ^vith a few of these
survivals of ancient Semitic paganism that I propose to deal in this
paper. At the outset it may be %vell to remind the reader that all
such legacies of the past are not equaUy worthy of condemnation.
Many of them are mere hannloss absurdities, or, if they have not
always been so, they have become so in the course of time, which has
gently stripped them of their harsher features, leaving behind what
is innocent and sometimes picturesque. These quaint survivals are
what we commonly mean by folk-lore; and accordingly it is of
some folk-lore elements in the Old Testament that I am about to
write. I shiill barely touch in passing on the darker and sadder
side of Semitic heathendom.

§ h The 3Iark of Cain.

We read in Genesis that when Cain had murdered his brother

Abel he was driven out from society to be a fugitive and vagabond

on earth. Fearing to be slain by any one who might meet him, he

remonstrated mth God on the hardness of his lot, and God had so

/ far compassion on liim that he * set a mark upon Cain, lest any man

N/ finding him should kill liim \^ What was the mark that God put on

^^the first murderer? or the sign that he appointed for him?

* 2 Kings xxiiL 4-24.

* Qeaesia iv. 8-15 (Authorized Version). The Revised Veraioo renders : * and
the Lord apfwinted a sign for Cain.* The most literal translation would be, * set
a sign to (or for) Cain,'

FOLK-LORE IN THE OLD TESTAMENT

103

That we have here a reminiscence of some old custom observed
by manslayers is highly probable ; and, though we cannot hope to
ascertain what the actual mark or sign was, a comparison of the
customs observed by manslayers in other parts of the world may
help us to understand at least its general significance, Kobertson
Smith thought that the mark in question was the tribal mark, a badge
which every member of the tribe wore on his person, and wliich
served to protect him by indicating that ho belonged to a community
that would avenge his murder J Certainly such marks are common
among savages. For example, among the Bedouins of to-day one of
the chief tribal badges is the mode of wearing the hair.^ In many
parts of the world, notably in Africa, the tribal mark consists of a
pattern tattooed or incised on some part of the person,^ That such
marks might serve as a protection to the tribesman in the way
supposed by Robertson Smith seems probable ; though on the other
hand it is to be remembered that in a hostile country they would,
on the contrary, increase his danger by advertising him as an enemy.

But even if we concede the protective value of a tribal mark,
still the explanation thus ofifered of the mark of Cain seems hardly
to fit the case. It is too general. Every member of a tribe was
equally protected by sucli a mark, whether he was a manslayer or
not. The whole drift of the narrative tends to show that the mark
in question was not worn by every member of the community, but
was peculiar to a murderer. Accordingly we seem driven to seek
for an explanation in another direction,

Fi'om the narrative itself we gather that Cain was supposed to
be obnoxious to other dangers than that of being slam as an outlaw
by any one who met him, God is represented saying to him: * What
hast thou done? the voice of thy brother's blood crieth unto me from
the ground. And now cursed art thou from the ground, which hath
opened her mouth to receive thy brother s blood from thy hand ;
when thou tillest the ground it shall not henceforth yield unto thee

' W, Eobertson Soiith, Kinship atid Marriage in Early Arabia ', p, 251.

' W- Robertson Smith, loc. cit,

^ J. G. Frazer^ Tokniismf pp« 28 sq. Thd evidence there adduced might be
indefinitely multiplied, especially for Africa. In the work to which I Imve referred
I waa mistaken in attempting to connect tribal marks with totemism. Probably
auch marks are seldom or never totemic^ aince they are common to all members of
a tribe ; whereas a totemic mark would be confined to one particular subdivision
(clan or ffc^^) of the tribe.

104

FOLK-LORE IN THE OLD TESTAMENT

her Btrength ; a fugitive and a wanderer shalt thou be in the earth/ ^
Here it is obvious that the blood of his murdered brother is regarded
as constituting a physical danger to the murderer; it taints the ground
and prevents it from yielding its increase. Thus the murderer is
thought to have poisoned the sources of life and thereby endangered
the supply of food for himself, and perhaps for others. On this view
it is intelligible that a homicide should be shunned and banished the
country, to which his presence is a continual menace. He is plague-
stricken, surrounded by a poisonous atmosphere, infected by a con-
tagion of death ; his very touch may blight the earth. Hence we
can understand a certain rule of Attic law. A homicide who had
been banished, and against whom in his absence a second charge had
been brought, was allowed to return to Attica to plead in his defence,
but he might not set foot on the land> he had to speak from a ship,
and even the ship might not cast anchor or put out a gangway. The
judges avoided all contact with the culprit, for they judged the case
sitting or standing on the shore,^ Clearly the intention of this rule
of law was to put the manslayer in quarantine, lest by touching Attic
earth even indirectly through the anchor or the gangway he should
blast it For the same reason, if such a man, sailing the sea, had
the misfortune to be cast away on the country where his crime had
been perpetrated, he was allowed indeed to camp on the shore till a
ship came to take him oflf, but he was expected to keep his feet in
the sea- water all the time^; evidently in order to counteractj or at
least dilute, the poison which he was supposed to instil into the soil.
Thus a mark put on a homicide might be intended primaiily
not for his protection, but for the protection of the persons who met
him ; it might be a danger signal to warn them off. If it was so, it
would serve at the same time indirectly to keep liim scathless.

^ OenesiB iv. 10-12 (Bevised Version).

* Demoathenes, xxiii 77 sq., pp. 645 sq. ; Aristotle, Constihttion of Athens, 57 j
Pausfuiias, i* 28. 11; Pollux, viiu 120; Helladius, quoted by Photius, Bibliotheca,
p. 535 a, lines 28 sqq., ed. I. 5ekker. The rule which forbade the ship to cast
anchor or to put out a gangway ia mentioned only by Pollux. But Pollux had
acceas to excellent authorities, and the rule bears the stamp of genuine antiquity*
We may therefore safely dismiss as unauthorized the statement of HeUadius that
the ship cast anchor.

* Plato, Laws, ix. 8, p* 866 cd. In ancient Greece, for a different reason, when
a man died of dropsy, his children were made to sit with their feet in water until
the body was burned (Plutarch, De sera numink vindicta^ 14). See my Lectures on
the Early History o/ihe Kingship^ p. 47.

FOLK-LORE IN THE OLD TESTAMENT

105

HoweveTj a closer examination of the danger which clung like
a plague to the manslayer may lead us to a different interpretation of
the murderer's mark. Here again, as in the customs just mentioned,
we seem to touch the bed-rock of superstition in Attica, Plato tells
us that according to a very ancient Greek belief the ghost of a man
who had just been killed was angry with his slayer and troubled him,
being enraged at the sight of the homicide going about in his old
familiar haunts ; hence it was needful for the manslayer to depart
from his country for a year until the wrath of the ghost had cooled
down^ nor might he return till sacrifices had been oflPered and cere-
monies of purification performed. If the victim chanced to be a
foreigner, the homicide had to shun the native land of the dead man
as well as his own, and in going into banishment he had to follow a
prescribed road \ for clearly it would never do to let him rove about
the country with the angry ghost at his heels. Among the Yaos
and perhaps other tribes of British Central Africa * the man who
kills his own slave, or even his younger brother or other ward, is not
amenable to justice, but — unless he can protect himself by a charm^ —
he is afraid of the mysterious chirope which overtakes those who
shed blood within the tribe. The chief, to whom he goes if he has
committed such a miu^der, procures the charm for liim from liis own
medicine-man, and uses it himself as well, ** because of the blood
that has been shed in his land/"^ The mysterious chirope which
thus overtakes a man who has shed blood within the tribe is ex-
plained to be either an illness or a sort of madness which comes
over him, as it is said to have come over Orestes after the murder of
his mother ^, until he has performed an expiatory ceremony ; and
* the idea is that the spirit of the slain enters into the body of the
slayer'. When the homicide has used the charm provided by
the chief, which may be either drunk or administered in a bath,
the danger passes away/

This fear of the wrathful ghost of the slain is probably at
the root of many ancient customs observed in connexion with
homicide ; it may well have been one of the principal motives
for inflicting capital punishment on murderers. For if such persons ^

* Plato, Laws, ix, 8, pp. 865 d-866 a ; DemostheneB, xxiii. pp. 643 sq. ;
HesyehiuB, s. v. dTrcvmi/ncr^d?.

' A. Werner, lite Natives of British Central Africa (London, 1906), p. 265.

* Pausanias, viiL 34. 1-4,

* A. Werner, op. cii,, pp* 67 Bq, ; Duff Macdonald, Afrlcana, i. 168.

106

FOLK-LORE IN THE OLD TESTAMENT

are dogged by a powerful and angry spirit which makes them
a danger to their fellows, society can obviously protect itself very
simply by sacrificing the murderer to the ghost, in other words
by putting him to death. But then it becomes necessary to guard
the executioners in their turn against the ghosts of their victims,
and this has been done, for example, by West African negroes ^
and some Indians of Brazil Among the latter people a man who
had publicly executed a prisoner had to fast and lie in his hammock
for three days, witliout setting foot on the ground ; further, he had
to make incisions in Ixis breast, arms, and other parts of his body,
and a black powder was rubbed into the wounds which left in-
eflfaceable scars so artistically arranged that they presented the
appearance of a tight*fitting garment. It was believed that he
would die if he did not observe these rules and draw blood from
his own body after slaughtering the captive.^ The fear of his
victim's ghost is not indeed mentioned by our authorities as the
motive for practising these customs. But that it was the real
motive is not only suggested by the analogy of the West Ahican
customs, but is practically proved by a custom which these same
Brazihan Indians observed before the execution. They formally
invited the doomed man to avenge his death, and for this purpose
they supplied him with stones or potsherds, which he hurled at his
guards, while they protected themselves against the missiles with

^ Gr. Loyar, in Aatley's Voyages and Travels, ii. 444 ; Father Bnudin, * Feti-
che urs ou minbtrea religieux dos N^gres de la Goin^e/ Missions Caiholiqties, xvL
(1884), p. 332 ; Major A G. Leonard, Jlie lower Niger and Us Tribes (London, 1906),
p. 180. According to Loyer the executioners are reckoned impure for three days
after an execution, and build a separate hut for themselves at a distance from the
village. There they live in seclusion for three days, after which they take the
hut to pieces^ leaving not so much as the aahea of their fire. Then *the lirst
executioner, having a pot on his head, leads them to the place where the criminal
suflfered. There they all call him thrice by his name. The first executioner breaks
his pot, and, leaving then* old rags and bundles^ they all scamper home.' According
to Father Baudinp the executioner at Porto Novo, on the coast of Guinea, used to
decorate his walls with the jawbones of his victims to prevent their ghosts from
troubling him at night.

* F. A. Theveti Lcs singularitez de la Framx Anturctique^ aufremefit nomm^
Amerique (Antwerp, 1568), p. 76; id.^ Cosmographie UniverseUe (Ptais, 1575), p. 946
[980] ; P. de MagaUianes do Gandavo, Histoire de la province de Sancta-Cniz (Paris,
1837)^ pp. 138 sq. ; T}te Captivity of Ham Stade of Hesse ^ London, Haklu>^ Society,
1874)j p. 159 ; J. Lary, Histona navigationis in Brasiliam quae el America dicitur
(1586), p. 192; R. Southey, History o/BrasU, l^ 232.

FOLK-LORE IN THE OLD TESTAMENT

107

shields made of hide.^ The form of the invitation, whicii ran thus,
* Avenge your death before your decease ', clearly implies a hope
that if the man had thus satisfied his thirst for vengeance in his
lifetime his ghost would not trouble them after death. But to
make assurance doubly sure the executioner secluded himself and
observed the curious precautions which I have described. The
drawing of blood from his own body, which was regarded as
essential to the preservation of his life '\ may have been intended
to satisfy the ghost's demand of blood for blood, while the permanent
marks left on the slayer's body would be a standing evidence tliat
he had given satisfaction to his victim* Could any reasonable
ghost ask for more ?

Tliis interpretation of the marks on the executioner s body is
conftmied by the following custom. Among the natives of New
Guinea, particularly near Finsch Harbour on the north-east coast,
the kinsmen of a murdered man who have accepted a blood-wit
instead of avenging his death take care to be marked with chalk
on the forehead by the relatives of the murderer, * lest the ghost
should trouble them for failing to avenge his death and should
carry off their pigs or make their teeth loose/ ^ In this custom
it is not the murderer but the kinsmen of his victim who are
marked, but the principle is the same. The ghost of the murdered
man naturally turns in fiiry on his unkind relatives who have not
exacted blood for his blood. But just as he is about to swoop
down on them to loosen their teeth, or steal their pigs, or make
himself unpleasant in other ways, he is brought up short by the
sight of tlie white mark on their black or coffeeKJoloured brows.
It is the receipt for the payment in full of the blood-wit : it has
been Uterally chalked up there by his own kinsmen : he cannot
truthfidly deny their signature : he is balked, and turns away dis-
appointed. The same mark might obviously be made for the same
reason on the murderer's brow to prove that he had paid in cash,
or whatever may be the local equivalent of cash, for the deed he
had done, and that the ghost therefore had no further claim on
him. Was the mark of Cain a mark of this sort ? Was it a proof
that he had paid the blood- wit ? Was it a receipt for cash down ?

It may have been so, but there is still another possibihty to

^ J* Lery, op. cit, p. 185.

' K de Magalhanes de Gatidavo, op. ciL, p. 139.

* B, Hagen, Untcr den Papuas CVVieabaden, 1899), p. 254.

108

FOLK-LORE IN THE OLD TESTAMENT

be considered. On the theory which I have just indicated it is
obvious that the mark of Cain could only be put on a homicide
when his victim was a man of the same tribe or community as
himself, since it is only to men of the same tribe or community
that compensation for homicide is paid. But the ghosts of slain
enemies are certainly not less dreaded than the ghosts of slain
friends ; and if you cannot pacify them with a sum of money paid
to their kinsfolk, what are you to do with them? Many plans
have been adopted for the protection of warriors against the spirits
of the men whom they have sent out of the world before their due
time. Apparently one of these precautions is to disguise the slayer
so that the ghost may not recognize him ; another is to render Iiis
person in some way so formidable or so offensive that the spirit
will not meddle with him. One or other of these motives may
explain the following customs, which I select from a large number
of similar cases.

Among the Ba-Yaka, a Bantu people of the Congo Free State^
* a man who has been killed in battle is supposed to send his soul
to avenge his death on the person of the man who killed him;
the latter, however, can escape the vengeance of the dead by wearing
the red tail*feathers of the paiTot in his hair, and painting his
forehead red/ ^ Among the Thompson Indians of British Columbia
it used to be customary for men who had slain enemies to blacken
their faces. If this precaution were neglected it was believed that
the spirits of their victims would bUnd them," Among the Angoni,
a Zulu tribe settled to the north of the Zambesi, warriors who have
killed foes on an expedition smear their bodies and faces with
ashes, hang garments of their victims on their persons, and tie
ropes rotmd their necks, so that the ends hang down over their
shoulders or breasts. This costume they wear for three days after
their return, and rising at break of day they run through the
village uttering frightful yells to drive away the ghosts of the slain,
which, if they were not thus banished, might bring sickness and
misfortune on the inhabitants.^ Among the Bantu tribes of Kavi-

' E. Torday aod T. A, Joyce, * Notes on the Ethnography of the Ba-Yaka,'
Jmimal o/tht Anthropological InstUutCj xxxvi (1906), pp. 50 &q.

" J, Teit, * The Thompeon Indians of British Columbia/ Memoirs of the
American Museum of Natural Bhtori/j vol ii. Anihr^ology, u [Part] iv* ([New
York,] April, 1900f, p. 357.

'^ C, Wiese, *Beitr%e zur Geschichte der Zulu im Norden dea Zambesi,'
Zciischrifi fur Ethnologit^ xxxii (1900), pp. 197 sq.

FOLK-LORE IN THE OLD TESTAMENT

109

rondo, in eastern Africa, when a man has killed an enemy in
warfare he shaves his head on his return home, and his friends
rub a medicine, which generally consists of goat's dung, over his
body to prevent the spirit of the slain man from troubling him.*
With the Ja*Luo of Kavirondo the custom is somewhat different.
Three days after his return from the fight the warrior shaves his
head. But before he may enter his village he has to hang a live
fowl, head uppermost, round his neck ; then the bird is decapitated
and ite head left hanging on his body. Soon after his return
a feast is made for the slain man, in order that his ghost may not
haunt his slayer.- In Fiji any one who had clubbed a human being
to death in war was consecrated or tabooed. He was smeared
red by the king with turmeric from the roots of his hair to liis
heels, A hut was built and in it he had to pass the next three
nights, during which he might not lie down, but must sleep as
he sat. Till the three nights had elapsed he might not change
his garment, nor remove the turmeric, nor enter a house in which
there was a woman. ^ That these rules were intended to protect
the Fijian warrior from his victim's ghost is strongly suggested,
if not proved, by another Fijian custom. When these savages had
buried a man alive, as they often did, they used at nightfall to
make a great uproar by means of bamboos, trumpet-shells, and so
forth, for the purpose of frightening away his ghost, lest he should
attempt to return to his old home. And to render his house un-
attractive to him they dismantled it and clothed it with everjrthing
that to their ideas seemed most repulsive,* So the North American
Lidians used to run through the village with hideous yells, beating
on the furniture, walls, and roofs of the huts to drive away the
angry ghost of an enemy whom they had just tortured to death. ^
A similar custom is stUl observed in various parts of New Guinea/

1 Sir H. Johnston, The Uganda Protectorate (London, 1902), iL 743 aq. ;
C, W, Hobley, Eastern Uganda (London, 1902), p< 20.

* Sir H. Johnston, op. cit, ii. 794 ; C. W. Hobley, op, cit., p* 31,

* T. WOliama, Fiji afid the Fijians ', i* 55 sq.

* J. E. Erskine, TJte Western Facifc, p. 477.

* Charlevoix, Histoire de la Nouvelle France, vL 77, 122 aq. ; J. F, Lafitau,
Momrs des Sauvages Amenquaim, il 279,

* R. E. Guise, in Journal of the Anthropological Institute, xxviii (1899),
pp. 213 sq. ; J. L. D. van der Roeat, in Tijdsehrift troor Indische Taal-^ Land- en
Volkenkund€y xJ (1898), pp. 157 sq, ; H. von Roaenherg, Der malayische Archipel
p, 461 ; K Vetter in NachriMen uber Kaiser Wilhelms-Land und deft Bismarck-
Archipel, 1897, p. 94,

110

FOLK-LORE IN THE OLD TESTAMENT

j Thus the mark of Cain may have been a mode of disguising
a homicide or of rendering him so repulsive or formidable in appear-
ance that his victim's ghosts would either not know him or at least
give him a wide berth. Elsewhere I have conjectured that mourning
costume in general was originally a disguise adopted to protect the
surviving relatives from the dreaded ghost of the recently departed.^
Whether that be so or not, it is certain that the living do sometimes
disguise themselves to escape the notice of the dead. Thus in the
western districts of Timor, a large island of the Indian Archipelago,
before the body of a man is coffined, his wives stand weeping over
him, and their village gossips must also be present * all with loosened
hair in order to make themselves unrecognizable by the nitu (spirit)
of the dead.' - Again, among the Herero of South Africa, when
a man is dying he will sometimes say to a person whom he does not
Uke, * Whence do you come ? I do not wish to see you here,' and
so saying he presses the fingers of his left hand together in such
a way that the tip of the thumb protrudes between the fingers.
* The person spoken to now knows that the other has decided upon
taking him][away {okuttmerera) after his death, which means that he
must die. In many cases, however, he can avoid this threatening
danger of death. For this purpose, he hastily leaves the place of
the dying man and looks for an onganga (i. e. ** doctor, magician"),
in order to have himself undressed, washed, and greased again, and
dressed with other clothes. He is now quite at ease about the
threatening of death caused by the deceased ; for, says he, ** Now
our father does not know me.*'*^ In like manner we may suppose
that when Cain had been marked by God he felt quite easy in liis
mind, believing that the ghost of his murdered brother would no
longer recognize and molest him*

§ 2. Sacred oaks and terebinths.

^\ Among the sacred trees of the ancient Hebrews the oak and

the terebinth seem to have held a foremost place. Both trees are
still common in Palestine. Thus, for example, speaking of the Plain

* Journal of the Anthropological In^tifut€j xv (1886), p. 73.

" J. G. F. Riedel, * Die Land&chaft Da wan oder Weat^Timor,' Deutsche Geo-
graphische BUtier, x. 286.

^ The Rev, G, Viehe, * Some customs of the Ovaherero/ (South Afrkan) Folk-
lore Journal^ i(1879), pp. 51 sq.

FOLK-LORE IN THE OLD TESTAMENT

111

of Sharon which is interposed between the inliospitable sandy shore
of the Mediterranean and the hills of Samaria, Thomson says : * The
sandy downs, with their pine bushes, are falling back towards the
sea, giving place to a firmer soil, upon which stand here and there
venerable oak trees, like patriarchs of by-gone generations left alone
in the wilderness. They are the beginning of the largest and most
impressive oak forest in western Palestine. It extends northwards
to the eastern base of Carmel, and^ with slight interruptions, it
continues along the western slopes of Galilee quite to the lofty
Jermuk, west of Safet, I have spent many days in wandering
through those vast oak glades. The scenery is becoming quite
park-like and very pretty. The trees are all of one kind, and
apparently very old. The Arabic name for this species of oak is
8indidn — ^a large evergreen tree whose botanical name is quercus
psetido-^occifera.^ There are other varieties of the oak interspersed
occasionally with these, but the prevailing tree everywhere is the
noble, venerable, and solemn sindidn, , . . On one occasion I spent
a night, for the sake of protection, at a village a few miles north-
east of these mills called Sindiflneh — the name no doubt derived
from the oak woods which surround it I had a delightful ramble
early the next morning in those grand old forests, and then under-
stood perfectly how Absalom could be caught by the thick branches
of an oak. The strong arms of these trees spread out so near the
ground that one cannot walk erect beneath them ; and on a frightened
mule such a head of hair as that vain but wicked son polled every
year would certainly become inextricably entangled/ ^ In antiquity
these woods of Sharon were known as the Forest or the Oak Forest,
and they are the Enchanted Forest of Tasso,^ Again, in speak-
ing of the Wady *Abilin on the confines of Zebulun and Asher,
Thomson says : * It is conducting us through a grand avenue of
magnificent oaks, whose grateful shade is refreshing to the weary

^ AmoDgst the many species of oaka found in Paleatine 'this variety is the
moei common, and sometimes attains a magnifioent growth, as the oak of Libbeya
in Galilee ' (H, B, Tristram, The Fauna and Flora of Palestine^ p. 412). Compare
id.f l%e Natural History of the Bible '^ pp. 368 sq. As to the oak of Libbeya, flee
below, p. 112;

' W. M. Thomson, T^ie Land and the Book, Southern Pakstine and Jerusalem
(London, 1881), pp. 60 sq. ; compare p, 79,

* G. A. Smith, The Historical Geography of the Holy Land (London, 1894),
pp. 147 sq.

112

FOLK-LORE 15 THE OLD TBHTTAMKHT

tntieikT. Tbej mre fart of ai
of the hiUs Miitliw»d to the piiin of Esdn^on. Thmm m hmrdfy
a more Bgnethle rida in the oocmtfy than througb tliiB noble oak
wood from Sbe& *Omar to Seffidnek. Many of the tnm are verr
Urgtj and bj tbeir great age indicate that tldia region waa not
eiiltivatod.' ' Again, tba romantir aceoorf off Biiiini^ wiiare the
JTovi&m boreta fitU-botn from IIib fiiot of Moont Hemiuii, tiwia much
of ita charm to foreata and dumps of grand oafcau* Omon Xkiatram
deaeiibea an ereigreen oak at the Tillage of Libbeja in thia
oeigfabourbood aa the most magnifioent tree he eror renambered to
have Men. At a little distance he and \m fneadB eontd hardlj
beliere that it waa a nngle tree.^

Paaring now to the east of the Jordan we are told of Ard el
Balfaanjehi the ancjent Batanea, that 'the whole of the province
ia eaeeedingly pi etare a q tie. The mountains are well wooded with
fereata of evatgre^i oeka, and the sides terraeed.* « Again, apeaking
of the Decapolig, Thomson writes : * We have been foOowing along
tibe nmatna of a Romaa road^ and now we are entering a beautiful
fiMreat of eveigreen oaks which seems to aYtiimi a graefe distance
over the range of Jebel Haurfta Etinawit itaKir ia avminded by
it, and many of the ruiujs are embowered beBe«& widai^raeding
ihididm trees, as these scrub-oaks are called by the natives, and
here and there some of the colunms are seeft nM|f above the dense
foliage'^; and further on he writes: ^The coimtry brtween our
line of travel and the valley of the Jordan northward and west-
ward is wild and mountainous, and in some parts it is well
wooded with noble oak forests. It is the region of the ancient

> W* M* TbomflOD, The Land and the Book, Omind MbtHm amti Photmieitk
p« 302* Am to this oak-forest see farther H. B. Tneitam^ Tki Land of Israd*.
pp« 112, 116, 12L However, since Thomaoo wrote, the destnietiaD of the foredU
of western Pale^ine would seem to have proceeded apAce. See BL B, Tristrmm,
The Natural HUiorp of ike Bible \ p. 7-

* W. M. Thoxuflon, The Land and the Book, Cenirui PaieMme and Fhoenkia,
pp. 440, 464, 467, 469, 470, 473, 481, 484, 485, 494 ; H. B. Tristram, The Land of
Israel^ pp. 572, 678, 577, 578.

* H. B. Trietranif op* cU,^ pp. 594 sq. The lower trunk of the tree measured
thirtjr-fteren feet in circumference at the narrowest part.

* Dr, Porter, quoted by W. M. Thomdon, The Land and the Book^ L^fanon,
I}ama$cu§t and fteytmd Jordan, p, 441.

^ W. H. Thomson, The Land and Ifie Book, Lebanon, Damascus, and be^fond
Jordan, p. 481 ; compare pp. 494, 497.

FOLK-LORE IN THE OLD TESTAMENT

113

Decapolis/ ^ ^ In Gilead, we come to a more densely-wooded region,
a true forest in places, the tops of the higher ranges covered with
noble pines ; then a zone of evergreen oaks^ with arbutus, myrtle,
and other shmbs intermixed; lower down the deciduous oak is
the predominant tree.' ^ Of these beautiful woods of Gilead, where
the famous balm was obtained, Thomson says: *We have now
reached the regular road from el Husn to Suf and Jerasli, and will
have the shade of this noble forest of oak, pine, and other trees for
the rest of the ride* There is not a breath of air in these thick
woods, and the heat is most oppressive both to ourselves and our
weary animals. , • . Up to this point — an hour and a half from
el Husn — much of the country is cultivated, but from this cm to
Suf the forest is uninterrupted » and is composed mostly of ever-
green oaks, interspersed occasionally with pines, terebinths, and
hawthorn, • , , From Um el Khanzir to Suf is nearly two hours,
and in spring nothing can be more delightful tlian a ride through
tliese forests, the grandest in this land of Gilead ; and we need not
wonder at the encomiums lavished by all travellers that have passed
this way on the beautiful woodland scenery of these regions, for
even the most enthusiastic have not said enough in its praise.'^
'After leaving the olive groves of Saf we shall be overshadowed
by an uninterrupted forest of venerable oak and otlier evergreen
trees for more than an hour to 'Ain-Jenneh. . . . These forests
extend a great distance to the north and south, and a large part
of the country might be brought under cultivation by clearing away
the trees. The substratum is evei-ywhere limestone, the soil is
naturally fertile, and in the spring of the year the surface is
clothed with luxuriant piiisture. " Jebel Ajlun", says Dn Eli Smith,
** presents tlie most charming rural scenery that I have seen in
Syria ; a continued forest of noble trees, chiefly the evergreen oak,
sindldn^ covers a large part of it, while the ground beneath is
clothed with luxuriant grass, a foot or more in height, and decked
with a rich variety of wild flowers," ' * Speaking of this district
Canon Tristram says : * Our second hour was through real forest,
by winding paths and under spreading oaks, where many a turban

» W. M. Thomson, op. ciL, p. 546.

* H. B. Trislram, The Xatuml History of the Bible ^, p* a
^ W. M. Thomson » op, cf/.» p. 555. Aa to the oak -woods of Gilead, see also

J. L, Burckhardt, Travchi in St/ria and (he Hofr/ Land (London, 1822), p. 348.

* W. M. Thomson^ op, rit., pp, 574 sq., compare p. 582,

T¥t4>M I

FOLK-LORE IN THE OLD TESTAMENT

was knocked off, or mule's burden dislodged.* * * Immediately
Ijeyond Khirbet Silr we began to descend into Wady es Seii- by
a very steep path, through a magnificent forest of large oak-trees.
That valley is very beautiful, imd the mountains rise liigher and
higher on either side, covered to their summits with thick groves
of evergreen oaks, terebinths, and other trees/- Not far off, in
a rocky amphitheatre commanding a wide prospect westward, and
backed on all other sides by wooded hills and jagged limestone crags,
are the ruins of the castle which Hyrcanus, one of the Maccabean
princes, built for himself, and adorned with spacious gardens, when
he retired in dudgeon to live in rural solitude far from the intrigues
and^tumults of Jerusalem, He was a wise man to choose so fair
a spot for his retirement from the world. The neighbouring glen,
the cliffs, the hill-sides wooded with oaks and terebinths, and the
green undulating slopes below make op a lovely landscape, especially
in spring when the oleanders convert the bed of the purling stream
into a sheet of rosy bloom, ^

The oaks which thus abound in many parts of Palestine are
still regarded with superstitious veneration by the peasantry. Thus,
speaking of a fine oak grove near the Lake of Phiala in northern
Palestine, Thomson says: 'These oaks under which w^e now sit
are believed to be inhabited by 3 An and other spirits. Almost eveiy
village in these wadys and on those mountains has one or mom
of such thick oaks, which are sacred from the same superstition.
Many of them in this region are believed to be inhabited by certain
spirits, called Bendt Ya'k6b — daughtei-s of Jacob — ^a strange and
obscure notion, in regard to which I could never obtahx an
intelligible explanation. It seems to be a relic of ancient idolatry,
which the stringent laws of Muhammed banished in form, but
could not entirely eradicate from the minds of the multitude.
Indeed, the Moslems are as stupidly given to such superstitions
as any class of the community. Connected with this notion, no
doubt, is the custom of burying their holy men and so-called
prophets under those trees, and erecting mumrs [domed shrines]
to them there. All non-Christian sects beKeve that the spirits of
these saints love to return to this world, and especially to visit the

^ H, B. Tristram, The Land of Israel \ p. 565.

* W. M, Thomson, op, at, p. 594.

• W* M. ThoniBonj op. cit, p. 596; H. B, Tristram, The Land of Im-aeV,
pp* 517 sqq- As to Hyrcaniifi and his castle see Josephus^ Antiquit, Jud, xii, 4. 11.

FOLK-LORE IN THE OLD TESTAMENT

115

place of their tombs. . • . I have witnessed some ludicrous disphiys
of daring enacted about such old trees by native Protestants just
emancipated from this superstition ; and I can point to many
people who have been all their lives long, and are still, held in
bondage through fear of those imiiginaiy spirits.

* Scarcely any tree figures more largely in biblical narrative *
and poetry than the oak ; but I observe that certain modern critics
contend that it is, after all, not the oak, but the terebinth. The
criticism is not quite so sweeping as that. It is merely attempted to
prove, I believe, that the Hebrew word ehth^ which in our vei'sion is
generally rendered oak, should be translated terebinth. Allon^ they
say, is the true name of the oak. The Hebrew writers seem to use
these names indiscriminately for the same tree or for diflferent'
varieties of it, and that tree was the oak. For example, the tree in
which Absalom was caught by the hair is called (Mh^ not the aUon ;
and yet I am persuaded it was an oak. Tlie battlefield on that
occasion was on the mountains east of the Jordan, always celebrated
for great oaks, I see it asserted by the advocates of this rendering
that the oak is not a common or very striking tree in this country,
implying that the terebinth is. A greater mistake could scarcely be
made. Besides the oak groves north of Tabor, and in Gilead, Bashan,
Hermon, and Lebanon, there are the forests, extending thirty miles
at least along the hills west of Nazareth to Carmel on the north,
and from there southward beyond Caesarea Palestina. To maintain,
therefore, that the oak is not a striking or abundant tree in Palestine
is a piece of critical hardihood tough as the tree itself,* ^

At the romantic viUage of Bludan, a favourite retreat of the
people of Damascus in the heat of summer, there are ' remains of
an old temple of Baal ; and the grove of aged oaks on the slope
beneath it is still a place held in superstitious veneration by the
villagers'.^ *In the TF, Barado^ near Damascus, where certain •
heathenish festival customs do yet remam amongst the Moslemin,
I have visited two groves of evergreen oaks, which are unshing-phwes
for the peasantry. If anytliing fall to them for which they vowed,
they will go to the one on a certain day in the year to break a crock
there ; or they lay up a new stean in a little cave which is under a
rock at the other. Thei-e I have looked in, and saw it full to the

* W. M. Thomson, Ifte Land and tfie Book^ Central Palestine and Fhoaiida,
pp. 474--6.

' H. B. Tiiatmm, Tlte Land afl$mel'\ p. 164.

I 2

116

FOLK-LORE IN THE OLD TESTAMENT

entry af tlioir yet whole offering-pots : in that other grove you will
see the heap of their broken potsherds/ ^ Another sacred grove of
oaks is at Beinu in northern Syria, A ruined Greek church stands
among the trees.- Again, we are told that *in a Turkish village
in northern Syria there is a large and very old oak-tree, which is
regarded as sacred. People burn incense to it, and bring their
offerings to it, precisely in the same way as to some shrine. There
Ls no tomb of any saint in ics neighbourhood, but the people worship
the tree itself'.^

Very often these venerated oaks are found growing singly or in
groves beside one of those whiteniomed tombs or supposed shrines
)f Mohammedan saints, which may be seen from one end of Syria

* to the other. Many such white domes and green groves crown the
- tops of liills, *Yet no one knows when, by whom, or for what
V /Special reason I hey first became consecrated shrines. Many of them

V are dedicated to tlie patriarchs and prophets, a few to Jesus and the

/ ai>ostlcs; some bear the name of traditionary heroes, and others

appear to honour pemons, places, and incidents of merely local

♦ interest. Many of these *'liigh places" have probably come down
from remote ages, through all the mutations of djTiasties and
reUgions, unchanged to tho present day. We can believe this the
more readily Ijecause some of them are now frequented by the oldest
communities in the country, and those most opposed to each other —
Arabs of the desert, Muhammedans, Metawileh, Druses, Christians,

I i and even Jews. We may have, therefore, in those *' high places

' \/ under every green tree upon the high mountains and upon the hills '',

/xnot only sites of the very highest antiquity, but existing monuments,

\y with their groves and domes, of man's ancient superstitions ; and if

tliat does not add to our veneration, it will greatly increase the

interest with which we examine them. There is one of these '* high

places", with its groves of venerable oak-trees, on the summit of

Lebanon, east of this village of Jezzin, The top of the mountain is

of an oval shape, and the grove was planted regularly around it/ *

In like manner Captain Condor, speaking of the real, not tho

nominal, rehgion of the Syrian peasantry at the present day, writes

* C. M, Doughty, Tramh in Arabia Dcserta (Cambridge, 1888), i. 450.

* S. I. Curtiss, Primitive Snnitic Eeligiofi To-dat/ (Chicago, 1902), pp, 138 sq.
' S. I* Curtifis, op. ciLy p. 94.

* W. M. Thomson, Tlie Land and the Book^ Lehamn, Damascus^ and hef^oml

Jordan, pp< 169-17L

FOLK-LORE IN THE OLD TESTAMENT

117

as Ibllows : ' The professed religion of the country is Islam, tho
simple creed of ** one God, and one messenger of God " ; yet you
may live for months in the out^ot-the-way parts of Palestine without
seeing a mosque, or hearing the call of the Muedhen to prayer. Still
the people are not witliout a religion which shapes every action of
their daily life. • . , Li almost every village in the country a small
building, sunnoimted by a white-washed dome, is observable, being
the sacred chapel of the place ; it is variously called Kubhehj '*dome" ;
Mazar^ ** shrine"; or Mukdm^ ''station"; the latter being a Hebrew
word used in the Bible for the "places" [mekomotk] of the Canaanites,
which Israel was commanded to destroy ** upon the high mountains,
and upon the hills, and under every gieen tree " (Deut. xii. 2). Just
as in the tmie of Mose^?, so now, the position chosen for the Mukdm
is generally conspicuous. On the top of a peak, or on the back of a
ridge, the little white dome gleams brightly in the sun ; under the
boughs of the spreading oak or terebinth, beside the solitary palm, or
:miong the aged lotus-trees at a spring, one lights constantly on the
low building, standing isolated or surrounded by the shallow graves
of a small cemeteiy. The trees beside the Mitlmms are always ^
considered sacred, and every bough which falls is treasured within
the sacred building,* ' This Mnhim represents the real religion uf
the peasant. , . , It is the sacred place from which the influence of
the saint is supposed to radiate, extending in the case of a powerful
Sheikh to a distance of perhaps twenty miles all round. If propitious,
the Sheikli bestows good luck, health, and general blessings on his
worshippers ; if enraged, he will inflict palpable blows, distraction
of mind, or even death, • • , When sickness prevails in a village,
votive offerings are brought to the Mukdm^ and I have often seen a
little earthenware lamp brought down by some poor wife or mother,
whose husband or child was sick. A vow to the saint is paid by a
sacrifice called Kod, or **requitar', a sheep being killed close to the
MtiMm and eaten at a feast in honour of the beneficent Sheikh/ '
Thus the worsliip at the high places and green trees wliich pious
Hebrew kings forbade and prophets thundered against thousands of
yeaiis ago persists in the same places to this day. So little is an
ignorant peasantry affected by the passing of empires, by the moral
and spiritual revolutions which change the face of the civilized
world.

To take, now, some particular examples of these local sanctuaries.
' C. R, Conder, TefU Work in Palesthie, ii. 218-22L

118

FOLK-LORE IN THE OLD TESTAMENT

On a ridge near the lake of Phiala in nortliem Palestine, there is
'a knoll covered witlx a copse of noble oak trees forming a truly
venerable grove, vnth a deep i eligious gloom \ In the midst of the
grove stands tlie welj/ or nlmne of Sheikh *Otliinan Haziiry ; it is
merely a common Moslem tomb surrounded by a sliabby stone walL
Ju&t below, on one side of the knoll, is a sm^ill fountain wliich takes

• its name from the saint. ' Again, on the summit of Jeljel Osh'a,
the highest mountain of Gilead, may be seen the reputed tomb of
the prophet Hosea shaded by a magnificent evergreen oak. The
tomb is venerated alike by Moslems, ChristianSj and Jews, People
used to come on pilgrimage to the spot to sacrifice, pray, and feast.
The prospect from the summit is esteemed the finest in all Palestine,
surpassing in beauty, though not in range, the more famous view
from Mount Nebo, whence Moses just before death gazed on the
Piomised Land, which he was not to enter, lying spread out in purple

If lights and sIukIows across the deep valley of the Jordan. ' Again,
the reputed tomb of Abel, high up a chff beside tho river Abana in
the Lebanon, is surrounded by venerable oak-trees. It is a domed
structure of the usual sort, and is a place of Mohammedan pilgrimage. *
At Tibneh a rockJiewn tomb is traditionally said to be the grave of
Joshua, iuid beside it grows a remarkable oak, which Captain Condei"

r describes as * perhaps the oldest and finest tree in Palestine \* Again,
at Tell el Kady, ' the hill of tho judge,' at the source of the Jordan,
a Moslem tomb is shaded by tw^o fine trees, a holm oak and a
terebinth standing side by side. Their branches are hung with rags
;md other rubbish, tho votive oiferings of pious people. '

Even when the hallowed oaks do not grow beside the tombs
or shrines of saints they are often thus decorated with rags by

* Edward Robinson, BiMical Jieamrdtcs in Pafcsthie^, iii, 401 ; W. M.
Thomson, The Land and the Book\ Cent ml Palestine and Pltoenkia, p. 473.

' J. L. Burckliardt, Trtweh iu Sf/ria and the Hofif Ijjnd [London, 1822),
pp. 853 sq. ; H. B. Tristram, The Land of Imtcl\ p, 546; W. M. Thomson, Thr
Land and tite Boah\ Lebanon j Damuscu^f and bcf/ond Jordan, pp. 585 sq. For the
view from Mount Nobu, see H, B. Tristram, Tlte Land of Israel \ pp. 624-7 ; id*.
The Land of 3Imh, i>p. 325 sq.

' W. M. Thomson, op. cit,, p. 350,

* W* M. Thomson, The Land and the Book^ Soullicrti Falestine and Jerusalem,
pp. rJl sq.

" H. B. Tristram, The I^nd of Israel % pp. 572 sq, ; W. M. Thomson, The Land
and the Book, Central Palest hte ami Phoenicia, p. 459 (who does not mention the
species of the trees). Baedeker speaks only of an oak {Palestine and Syria \ p. 251>).

FOLK-LORE IN THE OLD TESTAMENT 119

the superstitious peasantry. Thus at Seiiilrij the site of the ancient
Shiloh, ' is a large and noble oak-tree called BalOitat-lbrahim^
Abraham's oak. It is one of the '* inliabited trees '' so common in .
this country, and the superstitious peasants hang bits of rags on
the branches to propitiate the mysterious beings that are supposed
to " inhabit" it/^ 'Some distance back we passed a cluster of large *
oak-trees, and the lower branches of one of thom were hung with
bits of rag of every variety of shape and colonr. What is the
meaning of this ornamentation? That was one of the haunted ^
or ' inhabited trees ', supposed to be the abode of evil spirits ; and \
tliose bits of rag are suspended upon the branches to protect the
wayfarer from their malign influence. There are many such trees <
in all parts of the country, and the superstitious inhabitants are
afraid to sleep under them/ - One of these hamited trees may be
seen on the site of Old Beyrouth It is a venerable evergreen oak
growing near the edge of a precipice. The people hang strips of
their garments on its boughs, believing that it has the power to
cure sickness. One of its roots forms an arch above ground, and
through this arch persons who suflFer from rheumatism and lumbago
crawl to be healed of their infirmities. Expectant mothers also ci*eep
through it to obtain an easy delivery. On the 21st of September
men and women dance and sing all night beside the tree, the sexes
dancing separately. This oak is so sacred that when a shallow
sceptic dared to cut a branch of it his ann withered up/^

* W. M, Thomson, ThclAindandtfte Book, Central Palestine and Phoenicia, p. 104,
Of this custom, as practised in Syria, the late Prof, S» I, Cuiiiss writea : * There
are many trees* Jipart from slirines, whicli are believed to be possessed by spirits, to
whom vows and sacrifices are made. 4Such ti*ees are offeen hung with rags or bits
of cloth. It is not easy to determine the significance of the rags. Bome say they
are intended to l>e a constant reminder to the saint of the petition of the worshipper,
like a string tied around the finger : others tiiat the rag taken from the ailing \x>dy
of the suppliant, and tied to one of the branches, is designed to transfer tlie iliaeaa
of the person represented by fclie rugs to th« saint, who thus takes it away from the
sufferers and bears it vicariously himself. Sometimes the man who is ill takes a rag
from the ti-ee, as one tears ofT a bit of the pail from the cenotaph of the shnne^ and
carries it about on bis person, and so enjoys tlie advantage of virtue from the
ssiint* (Prmitivt Semitic lieligioti To- da t/, p. 01). The custom of hanging rags on
sacred trees is observed in many hmds, though the motives for doing so are by no
means always clear. See E. S. Hartland, The Lcgmd of Perseus, il t75sqq.

' W. M, Thomson, op. cit, pp. 171 sq.

' F, Sessions, *Some Syrian folklore not^s gathered on Mount Lebanon/ Folk-
lore, ix (1898), pp. 915 sq. ; W, M. Thomson, O}). ciL, p. 190.

120

FOLK-LORE IN THE OLD TESTAMENT

In various paiis of the upper valley of the Jordan tliere are
groves of oaks and shrines dedicated to the daughters of Jacob.
One of these shrines may be seen at the town of Safed. It is
a small mosque containing a tomb in which the damsels are sup-
posed to live in all the bloom of beauty. Incense is burnt at the
door of the tomb, A gallant and now highly distinguished officer,
engaged in the Survey of Palestine, searched the tomb cai*efully
for the ladies, but without success J The association of the daugh-
ters of Jacob with oak-trees seems to point to a belief in Dryads
or nymphs of the oak.

In Hebrew the words commonly rendered *oak* and * tere-
binth * are very similar, the difference between them being in part
merely a dififerenee in the vowel points which were added to the
text by the Massoretic scribes in the Middle Ages. Scholars are
not agreed as to the coiTect equivalents of the words, so that w^hen
we meet with one or other of them in the Old Testament it is
to some extent doubtful whether the tree referred to is an oak or
a terebinth.'- The terebinth {Pistacia Terebinflms) is still a common
tree in Palestine, occurring either singly or in clumps mingled with
forests of oaL The natives call it the htfm tree. It * is not an
evergreen, as is often represented ; but its small feathered lancet-
shaped leaves fall in the autumn and are renewed in the spring.
The flowers are small and fallowed by small oval hemes, hanging
in clusters from two to five inches long, resembling much the
clusters of the vine when the grapes are just set. From incisions
m the trunk there is said to flow a sort of transparent balsam,

' W, M. Thomson^ op. citf pp. 222» 445 sq. See also aboye^ p. 114.

■ 'There are five simiJar Ileb. worda — 'd [only in the pi. *iUm], 'cWi, 'don,
*(}Uah (only Jos. xxiv. 2^\ and 'allon — the difference between which depends in part
only upon the punctuation, and the specjul sense of each of which is not perfectly
certain : Gesenius, after a careful survey of tlie data, arrived at the conclusion,
which has been largely accepted by subsequent scholars, that 'e/, UliiJt, 'don denoted
properly the tcrehtntk, and *tdlah, 'Mon the oaJc * (S. R» Driver, TJw Book of Genesis *,
p, 147), See further Enajdopaedla BMica, a, v. * Terebinth". In regard to the
words in question Professor G. F. Moore maintains that * there is no real founda-
tion for the discrimination; the words signify in Aramaic *Uree " simply ; in
Hebrew usually, if not exclusively, '^holy tree," as the place, and primitively
the object^ of worship, without regard to the species ' {Commentary on Judges \
pp. 121 sq,). Canon Tristram held that 'eklh denoted the terebinth, but that
all the other words in question applied to acorn- bearing oaka According to him,
*alUn probably stands for the evergreen oak, and 'et^n for the deciduous sorts {The
Natuml Ilistorif of the Bilk *, p. 367)*

FOLK-LORE IN THE OLD TESTAMENT

121

constituting a very pure and fine species of turpentine, with an
agreeable odour like citron or jessamine and a mild taste, and
hardening gradually intt> a transparent gum. In Palestine nothing
seems to be known of tliis product of the Butm/ ^ The terebinth
* is a very common tree in the southern and eastern parts of the
country, being generally found in situations too wann or dry for
the oak, whose place it there supplies, and wliich it much resembles
in general appearance at a distance* It is seldom seen in clumps
or groves, never in forests, but stands isolated and weird-hke in
some bare ravine or on a hill-side, wliere nothing else towers above
the low bnishwood* When it sheds its leaves at the beginning
of winter, it still more recalls the familiar English oak, with its
short and gnarled trunk, spreading and irregular limbs, and small
twigs , . . Towards the north this tree becomes more scarce, but
in the ancient Moab and Anmion, and in the region mund Heshbon,
it is the only one which relieves the monotony of the rolling downs
and boundless sheep wallcs/ - Fint^ specimens of the tree may be
seen standing solitary in various places, for example one in the
Wady es Sunt on the way from Hebron to Ramleh, another at
the north-west corner of the walls of Jerusalem, another on the
supposed site of the city of Adidlain, and another at Shiloh, * And
beautiful forests of mingled terebinths and oaks clothe some of
the glens of the Lebanon, the hills of Naphtali and Galilee, and
form a great part of the rich woodlands on tlie eastern side of
the Jordan.^

Yet if we may judge from the comparative frequence of allu-
sions to the two trees in the descriptions of travellers, the terebinth

^ Edward I^binson, Biblieal liesearvhcjs in iWcv/ttia', ii. 222 sq* Compare
W. M* Thomson, The Land and (he Booh, Cmtml Palestine and Phocmcia, pp. 19 sq,,
who uIbo says that the resin is not extracted from the tree by the natives of
Palestine.

' H. B. Tristram, Tlie Natuml Hisionj of the Bihk\ pp. 400, 401.

' E, Kobinson, loc, cit. ; W. M, Thomson, TJic Land and the Book, Southern
Palestine and Jermalem, p. 229; id. Central Palest ine and Phoenicia j pp* 19 sq., 49 gq.,
478; H. B. Tristram, The Land of Ist^ad\ p. 159,

* W. M. Thomson, Tfie Land and the Book, Central Palestine and Phoenicia,
pp. 224, 257» 324, 551, 558, 559 ; id, Lebanon^ Damascus, and beyond Jordan,
pp. 282, 295, 502, 555. 578, 694, 596, 604 sq. See above, pp. 1 1:3, H 4/ On the road
from Heshbon to Rabbath Ammon * we rode np a narrow glen, rocky and rough,
with fine terebintli-trees, the largest we saw in Palestine, stretching their gnarled
and twisted boughs over the path ' (H. B, Tristram, The iMnd 0/ Israel \ p. 531),

122

FOLK-LORE IN THE OLD TESTAMENT

is less common in Palestine than the oak *, and is far lass often
the object of superstitious regard. Canon Tristram indeed tells
us that * many terebinths remain to tliis day objects of veneration
in their neighbourhood ; and the favourite bm-yiiig^place of the
Bedouin slieikh is under a solitary tree. Eastern traveller will
recall the *' Mother of rags" on the outskirt of the desert — a tere-
biutli covered mth the votive offerings of supei^stition or affection'^' ;
and elsewhere the same writer juentions a terebinth Ining with
rags at the source of the Jordan,^ Again, Captain Condor writes
that 'among the peculiar religious institutions of the comitry are
the sacred trees, which are generally oaks, or terebinths, with names
taken from some Sheikh to whom they t>elong. They are covered
aU over with rags tied to the branches, whicli are considered
acceptable offerings/^ But apart from these few notices (which,
however, might doubtlass be multiplied by further search)^ I have
found no evidence of a supei'stitious regard paid to the terebinth
by Syrian peasants in modern times. The rarity of such notices
compared with the abundant references to the sanctity of the oak
seems to show that in Syria at the present day the oak is more
^ commonly revered by the people than the terebinth ; and wlien
wc consider the tenacity and persistence of identical forms of
superstition through the ages we seem justified in concluding that
in antiquity also the oak was more generally worshipped than
the terebinth by the idolatrous inhabitants of the land. From
this it follows that when a doubt exists as to whether m the
Old Testament the Hebrew word for a sacred tree should be
rendered * oak ' or * terebinth ' the prefei^nce ought to be given
to the rendering ' oak '. This conclusion is confirmed by the general
practice of the old Greek translatoi-s and of St* Jerome, who, in
translating these passages, commonly render the doubtful word by
*oak' and not by 'terebinth','^ On the whole, then, the Revisers

^ €om{>ai^ tlie number of the mferences to oftks and terebinths respectively in
the indices to W. M. Thomson's Tfm Land and the Booh (the edition in three
volumes). From that work I have adduced only part of the evidence for the
prevalence of the oak. but most of the evidence for the prevalence of the terebinth*
No modern writer, probiibly, has known Syria and Palestine so well as Thomson,
who spent forty -live years of his life in the country*

^ H, B. Tristram, The Nahtml Histonf of the Bible% p. 401,

* See aljove, p. 1 18-

* C. R. Conder, Tent Work in Pctksihe, ii. 233,

* So far as I see, there are some eighteen to twenty passages in the Old

FOLKLORE IN THE OLD TESTAMENT

128

of our English Bible have done well to transliite all the words
ill question by ' oak ' instead of by ' terebinth ', except in the tw^o
passages where two of these words occur in the same verse. In
these two passages ' the Revisers render 'allon by * oak ' but 'eMi
by 'terebinth'. Elsewhere they render 'eldh by *oak', but in the
margin they mention ' terebinth ' as an alternative rendering. I shall
follow their example and cite the Revised Version in the sequel.

That the idolatrous Hebrews of antiquity revered the oak-tree
is proved by the evidence of the prophets who denounced the
superstition. Tlius Hosea says : * Tlioy sacrifice upon the tops of
the mountains, and burn incense upon the hills, under oaks and
poplars and terebinths, because the shadow thereof is good : there-
fore your daughters commit whoredom, and your brides commit
adulteiy. I will not punish your daughtei-s when they commit
whoredom, nor your brides when they commit adultery; for they
themselves go apart with whores, and they sacrifice Avith the
harlot^/ - The prophet here refei-s to a custom of reUgious prostitu-
tion which was carried on under the shadow of the sacred trees,
Referring to the sacred groves of his heathenish country^men,
Ezekiel says : ' And ye shall know that I am the Lord, when their

Testament where a reference is made to an oak or a terebinth, wMch, from the
context, may be thought to have heen sacrecl. In thirteen of these passages the
Septuagint renders the doubtful word by * oak * {Spt^ or fidkavos:}, and in iive by

* terebinth ' ; in the other passages the rendering is neutral. In eleven out of the
eighteen to twenty i>a.ssages St. Jerome, in his Latin version (the Vulgate),
rendei-s the doubtful word by ' oak * {qu€tTtis)f and in fom- by * terebinth * ; tn the
other passages the rendering is neutrab The passages in question are Genesis
xii, 6, xiii. 18, xiv. 13, xviii- 1, xxxv. 4 and 8 ; Deuteronomy xi. 30 ; Joshua
xxiv. 26 ; Judges vi* 11 and 19^ ix. 6 and 37 ; 1 Samuel x. 3 ; I Kings xiii. 14 ;
1 Chronicles x* 12 ; Isaiah i. 29 , lyii. 5 ; Jeremiah ii. 34 (where the Hebrew text
should be corrected by the Septuagint and the Peshitto ; see below, p. 124, note 4) ;
Ezekiel vi» 13 ; Hosea iv. 13. In a nimiber of thaae passages the English Authorized
Version is quite incorrect, rendering the doubtful word neither by * oak ' nor

* terebinth '. The Engliali reader should consult the Revised Version. In two passages
(Isaiah vi. 13; Hosea iv. 13) t^vo of the doul^tful words (dah and *a{l6n) occur in the
same verse. In the former passage the Septuagint renders 'clah by * terebinth' and
*alJon by *oak' (fftiXavo^:); in the latter passage it renders *allon by *oak' and *ehdi by

* shady tree '- In both passages the Vulgate renders 'dalt by * terebinth ' and ^allon
by *oak'. My ignorance of Syriac prevents me from comparing the renderings of
the Peshitto. I have to thank Professor F. C. Burkitt for kindly communicating
to me the rendering of the Peshitto in Jei'emiah ii. 34.

* Isaiah vi. 18 ; Hoaea iv. 13, See the preceding note.

• Hosea iv. 13sq.

124 FOLK-LORE IN THE OLD TESTAMENT

dlain men shall be among their idols rouud about their altars, upon
every high hill, in all the tops of the mountains, and under every
greeu tree, and under every thick oak, the place where they did
offer sweet savour to all their idols.' ^ Again, Isaiah, speaking uf
the sinners who forsake the Lord, says : * For they shall be ashamed
of the oaks which ye have desired, and ye shall be confounded for

* the gardens that ye have chosen/ - Again, the author of the later
prophecy which passes under the name of Isaiah, in denouncing the
idolatiy of his day, says : * Ye that inflame yourselves among the
oaks, under every green tree ; that slay the children in the valleys,

• under the clefts of the rocks/ ' The slaughter here referred to is no
doubt the sacrifice of children to Moloch. Jeremiah alludes to the
same practice in a passionate address to sinful Israel : * Also in thy
skirts is found the blood of the souls of the innocent poor : I have
not found it at the place of breaking in, but upon every oak/^

• Thus it would seem that the blood of the sacrificed children was
snieamd on, or at least offered in some form to, the sacred oaks.
In this connexion it should be remembered that the victims were
slaughtered before being burned in the fire% so that it would be

V possible to use their blood as an unguent or a hbation. The Gallas
of East Africa pour the blood of anunals at the foot of their sacred
trees in order to prevent them from witheiing, and sometimes they

^ Ezekiel vL 1B> For * oak * the Revised Version has * terebinth ' in the
margin.

* Isaiah i. 29* For * oaks ' the Eevised Version haa * terebinths' in the
margin «

^ Isainh Ivii. 5* For * among the oaks ' the Revised Version has * with idols *
in the margin. But the former rendering (or * among the terebinths *) is to be
preferred. See Professor J. Skinner in liis note on the passage {Isaiah xl-Ixvt^
p, 155, in the Camhridgc Bible for Scliools and Colleges),

* Jeremiah ii, 84, where the meaningless ^^^ (* these") of the Hassoretic
text should be corrected into n?K or npK (*oak ' or * terebinth 'J in accordance with
the readings of the Septuagint {iwl Tra<rji Spvt), and of the Syriac Version, Thi^
change ia merely one of punctuation ; the original Hebrew text remains unaffected.
The vague sense of the preposition /V leaves it uncertain whether the blood was
smeared on the trees or poured out at their foot. However, Professor Kennett
writes to me that he believes the textual corruption in Jeremiah ii. 84 to be too
deep to he healed by the alight emendation I have adopted. He conjectures that
the last clause of the verse is defective through the omission of a word or words,

* Genesis xxii; Ezekiel xvi* 20 sq,, xxiii. 39 ; G, F. Moore, in Emydopardia
Bihlica, iii, 3184 m^*

FOLK-LORE IN THE OLD TESTAMENT

125

smear the trunks and boughs with blood, butter, and milk.' In like '
manner tlie old Prussians sprinkled the blood of their sacrifices on
the holy oak at Remove ^, and Lucan says that in the sacred Druidical
grove at Marseilles every tree was washed with human blood. ^

But if in the later times of Israel the worship of the oak or the '
terebinth was denounced by the prophets as a heathenish rite, there
is a good deal of evidence to show that at an earher period sacred
oaks or terebinths played an important part in the popular religioii,
and that Jehovah himself was closely associated with them. At all
events it is remarkable how often God or his angel is said to have
revealed himself to one of the old patriarchs or heroes at an oak
or a terebinth. Thus the first recorded appearance of Jehovah to
Abraham took place at the oracular oak or terebinth of Shechem,
and there Abraham built Mm an altar.* Again, we are told that
Abraliam dwelt beside the oaks or terebinths of Manire at Hebron,
and that he built there also an altar to the Lord/'^ And it was
there, beside the oaks or terebinths of Mamre, as he sat in his tent
in the heat of the day, that God appeared to him iix the likeness
uf three men, and there under the shadow of the trees the
Deity partook of the flesh, the milk, and the curds which the
hospitable patriarch offered him.*^ So too the angel of tlie Lord
came and sat under the oak or terebinth of Ophrali, and Gideon,
who was busy threshing the wheat, brought liim the flesh and broth
of a kid and unleavened cakes to eat mider the oak. But the angel,
instead of eating the food, bade Gideon lay the flesh and cakes
on a rock and pour out the broth ; then with a touch of his staff
he drew fire from the rock, and the flame consumed the flesh
and the cakes. After that the heavenly, or perhaps the arboreal,

* Ph. Paulitschke, Ethnographic Nordost-AfrikaSf die geisiige CuUur dcr Damkil^
GaUa und Somdl (Berlin, 1896), pp. 34 sq. ; ui, Eihmgmphk Nordosi'Afnkas^ die
matericUe Cultur tier Bandkiif Galla und Sonidl (Berlin, 1893), p. 152. Compare
O. Baumanii, Usamhara und seine Nachhargchide (BerliD, 1891)^ p. 142,

' Hartknoeh, Alt und Nt^tes rrcussen (Frankfort and Leipaic, 1684), p. 159.

* Lucan, Fharsalia, iiL 405.

* Genesis adi* 6-9, The ' oak of Moreh ' (Revised Versioa, ' terebinth ' margin)
is the * directing oak ' or * oak of the director ' j where the reference is to oracular
direction given either by the tree itself or by the priests who served it Oracular
oaks or terebinths (oaks or terebinths of Moreh) are mentioned also in this neigh*
hourhood by the author of Deuteronomy (xi. 30)* See Professor S. R. Driver, 'fJw
Book of Gtnesis*, pp» 146 sq. ; id,, Cmimentan/ on Dcuterommg'^, p. 134.

* Geneeis xiii. 18, xiv. IS.

* Genesis xvin, 1-8, with Professor Driver's note on verse 8.

126

FOLK-LORE IN THE OLD TESTAMENT

visitant vanished, and Gideon, like Abraham, built an altar on the
spot.^

^ There was an oracular oak or terebinth near Shechem as well

as at Mamre^; whether it was the same tree under which God
appeared to Abraham, we do not know. Ita name, 'the oak or
terebinth of the augurs', seems to show that a set of wizards or
Druids, if we may call them so, had their station at the saci*ed
tree in order to interpret to inquirei-s the rustling of the leaves in
the wind, the cooing of wood-pigeons in the branches^ or such other

' omens as the spirit of the oak vouchsafed to his woi*shippers. The
beautiful vale of Shechem, embosomed in oHves, orange-groves, and
palms, and watered by plenteous rills, still presents perhaps the
richest landscape in all Palestine ', and of old it would seem to have
been a great seat of tree-worship. At all events in its history we
meet again and again with the mention of oaLs or terebinths wliich
from the context appear to have been sacred. Thus Jacob took the
idols or ^ strange gods ' of his household, together with the earrings
which had probably served as amulets, and buried them under the
oak or terebinth at Shechem.^ According to Eustathius the tree
mtm a terebinth and was worshipped by the people of the neighbour-

' hood down to his own time. An altar stood beside it on which

s sacrifices were offered*^ Again, it was under the oak by the sanctuai^
of the Lord at Shechem that Joshua set up a great stone as a witness,
saying to the Israelites, ' Behold this stone shall be a witness against
us ; for it hath heai'd all the words of the Lord which he spake unto
us : it shall be therefore a witness against you, lest ye deny your

^ Judges vL 11-24.

* Judges ix. 87, * the oak of Meoiienim ' (Revised Version), ' the augurs' oak
or terebinth" (Revised Version, margin). Compare G. F, Moore, (hmtncnianf on
Jmhjcs \ p. 260- We read of a man of God sitting under an oak (1 Kings xiii* 14) j
but the tree need not have been oracular*

* H* B. Tristram, The Land of Ismel\ pp, 1S5, 147, The modern name of
Shechem is Nablous. The town * has the mulberry, the orange, the pomegranate,
and other trees growing amongst the houses, and wreathed and festooned with
delicious perfume during the montlis of April and May, There the bulbul delights
to sing, and hundreds of other birds unite to swell the chorus. The people of
Nablus maintain that theirs is the most musical valley in Palestine, nor am I
disposed to contradict them ' (W, M, Thomson, The Land and the Book^ Ccnind
IkUcsthie ami Phoenicia^ p, 143),

* Genesis xxxv. 4, with Professor B. R. Driver's note.

^ Eustathius, quoted by H. Reland, Faiuesiina, p. 712.

FOLK-LORE IN THE OLD TESTAMENT

127

God.' ' And it was at * the oak of the pillar ' in Shecheni that the
men of the city made Abunelech king.- The oak or terebinth may
have been supposed to stand m some close rehition to the king ; for
elsewhere we read of a tree called * the king's oak ' on the Ijordei-s of
the tribe of Asher ^ ; and according to one account the bones of Saul
and of his sons were buried under the oak or terebinth at Jabesh^ So
when Rebekah's nui^se Deborah died, she was buried below Bethel
under the oak, and hence tlie tree was called the Oak of Weeping.''
The Oak of Weeping may perhaps have been the veiy oak at which,
according to the directions of Samuel the prophet, Saul shortly
before his coronation was to meet three men going up to sacrifice to
the Lord at Bethel, who would salute him and give him two of their
loaves.*^ This salutation of the future king by the three men at the
oak reminds ns of the meeting of Abraham with God in the likeness
of three men under the oaks of Mamre, In the original story the
greeting of the three men at the oak may have had a far deeper
meaning than transpires in the form in which the naiTativo has
come down to us. Taken along with the coronation of Abimelech
under an oak it suggests that the spirit of the oak, perhaps in triple
form, was expected to bless the king at his inauguration. In tlie
light of this suggestion the burial of Sauls bones under an oak
seems to acquire a fresh significance. The king, who at the
beginning of his reign had been blessed by the god of the oak,
was fittingly laid to his last rest under the sacred tree.

But of all the holy trees of ancient Palestine by far the most
famous and the most popular was appaiently the oak or terebinth
of Manire, where God revealed liimseLf to Abraham, the founder
of the Israelitish nation, in the likeness of three men. Was the
tree an oak or a terebinth ? The ancient testimonies are conflicting,
but the balance of evidence is in favour of the terebinth* ' Josephus
tells us that in his day many monuments of Abraliam, finely built

' Joshua xxiv- 20 sq.

* Judges ix. 6 ('terebinth/ Eeviaed Version, margin).
' Joshua xix. 26, where Allanielecli means * the king's oak \

* 1 Chroniclea x. 12. According to another account (1 Samuel xxxi* 8) the
tree under which the royal bones were buried was a tamarisk.

^ Genesis xxxv. 8.

•^ 1 Samuel x. 8,

'' The passages of ancient writero which refer to the tree are collected by
H. Reland, PaktesHna^ pp. 711-lo, and )>y Valesius in his commentar)^ on Euaebius,
ViL Ck>nstafUinif iil 53 (Wigne's Pairologia Graeca^ xx. lll^sqq.)*

128

FOLK'LORE IN THE OLD TESTAMENT

of beautiful marble^ wei-e shown at Hebron, and that six furlongs
from the town grew a very large terebinth, which w^as said to
have stood there since the creation of the world. ' Tliough he does
not expressly say so, we may assume that this terebmth was the
one under which Abraham was believed to have entertained the
angels. Again, Eusebius affirms that the terebinth remained down
to his own time in the early part of the fom*th century a.d,, and
that the spot was still revered as divine by the people of the
neighbourhood. A holy picture represented the three mysterious
guests who partook of Abraham s hospitality under the tree ; the
middle of the three figures excelled the rest in honour, and him
the good bishop identified with * Our Lord himself, our Saviour,
whom even they who know Him not adore \^ All three angels were
worshipped by the people of the neighbourhood.^ They curiously
remind us of the three gods whose images were worshipped in
the holy oak at Romove, tlie religious centre of the heathen
Pioissians,* Perhaps both at Hebron and at Komove the tree-god
was for some reason conceived in triple form. A pilgrim of
Bordeaux, author of the oldest Itincrm'y of Jerusalem^ writing in
the year 333 a.d*, tells us that the terebinth was two miles from
Hebron, and that a fine basilica had been built there by order of
Constantine. Yet from the manner of his reference to it we gathei-
that * the terebinth * was in his time merely the name of a place,
the tree itself having disappeared.^ Certainly Jerome, wanting later
in the same century, seems to imply that the tree no longer existed.
For he says that the oak of Abraham or of Mamre was shown down
to the reign of Constantine, and that * the place of the terebinth '

^ Josephus, BelL Jud^ iv. 9, 7.

' Eusebius, Demon^tratm Evangelicat v. 9 (M.igneB Patrolofjfia Gracca^ xxii, 384).
In his Onomasticon Eusebius, speaking of Hebron, mentions both the oak of
Abraham and the terebinth : t) 3pv? 'A/?paa/i., Koi to /AyiJ/ia aih-o^t Bimpurai^ icai
Bffrp-Krutrai tTnifntvCt^ irpo% rwv l)(0p<iiv [sicj rj $€p€J3tv$o^ teal ol toI 'A^paap. CTrcffi^tu^cvTc?

ayyeXoi (Eusebius, OnomasticoHf s. V. 'Ap^ai, pp. 54, 56, od. F. Larsow and
Gr, Parthey). In this passage we must read 7rkija-Lox*^p<^Vf or iyxiMipim^ or some such
word for ixj^fmi.

• Eusebius, Onomasticon, s. v* 'Ap^u'u See the preceding note.

* Hartknoch, Alt und Neues Preussen (Frankfort and Leipsic, 1684), pp. 116sq.
^* Mtinerarium Burdigalense *, in Itinera Hierosoli^fniiana, rec. P. Geyer

(Vienna, 1898), p. 25: Inde Tcrebintlio milia mii* Ubi Abraham habHavU et
ptitmm fodit stih arbore tcrebintho et cum angelis locutus est et cibum sumpsitf ihi
basilica facta est jui^m Comianthii viirae pukhritudinis, hide terebintho Cebron
milia ih

FOLK-LORE IN THE OLD TESTAMENT

129

was woi-shipped superstitiously by all the people round about,
because Abraliam had tliere entertaiued the angels.*

When Constantine deteniiined to build a church at the sacred
treej he commumcated his intention in a letter to Eusebius, bishop of
Caesarea, who has fortunately preserved a copy of the letter in Ma
life of the emperor. I will extract from it the passage which relates
to the holy tree ; ' The place which in called ** at the Oak of Mamre ",
where we learn that Abraham had his home, is said to be polluted
by cei-tain superstitious persons in various ways ; for it is reported
that most damnable idols are set up beside it, and that an altar
stands hard by, and that unclean sacrifices are constantly offered.
Wherefore, seeing that this appears to be foreign to the present
age and unworthy of the holiness of the place, I wish your Grace
to know that I have written to the right honourable Count Acacius, my
friend, commanding that without delay all the idols found at the afore-
said place shall be committed to the flames, and the iiltar overturned ;
and any one who after this decree may dare to commit impiety in such
a place shall be deemed liable to punishment We have ordered
that the spot shall be adorned with the pure building of a basilica,
in order that it may be made a meeting-place worthy of holy men/ ^

In this letter it will be observed tliat the emperor speaks of
the sacred tree as an oak, not as a terebinth, and it is called an oak
also by the Church liistoiians Socrates ^ and Sozomenus \ But little
weight can be given to their testimony, since all three probably
followed the reading of the Septuagint, which calls the tree an
joak, not a terebinth,^ It is probably in deference to the authority
^f the Septuagint that Eusebius himself speaks of "the oak of
Abraham " in the veiy passage ui which he teUs us that the terebinth
existed to liis own time."^ The Church historian Sozomenus has
bequeathed to us a curious and valuable description of the festival

* Jerome, Liber de situ et nominibus loconim Ilebtuicorum, s. v. * Arboc '
(Migne's Patrologia Latina, xxiii* 862), This treatise of Jerome, which is sub-
stantially a translation of the Onomustkon of Eusebius^ was written about 388 a. D.
It 19 printed in the convenient edition of the latter work by Larsow and Parthey.

- Eusebius, Vit. Constantini, iii. biS (Mignes Patrologia Grtteca, xx. ni2aqq.).

* Socrates, Historia Ecclcikistlca^ L 18 (Migne's Patrologia Graeca, Ixvii. 124),
who seems to draw his information from Eusebius s Life of Constantine.

* Sozomenus, Historia Ecclesiastical ii. 4 (Migne s Patwlofjia Graeca, Ixvii. 941,
4144). Tet while he speaks of ' the oak called Mamre \ this historian tells us that

/ the place itself was called Terebinth,

^ Genesis xiii* IS, xiv. lo, xviii, !♦ * See above, p, 128, note 2.

TITLOU K

130

FOLK-LORE IN THE OLD TESTAMENT

which down to the time of Const^antinej or even later, was held every
Slimmer at the sacred tree. His account runs thus :

' I must now relate the decree which the Emperor Constantino
passed with regard to what is called the oak of Manire. This place,
which they now call Terebinth, is fifteen furlongs north of Hebron
and about two hundred and fifty fiirlongs from Jerusalem. It is
a tine tale that with the angels sent against the people of Sodom the
Son of God appeared to Abraliam and told him of the birth of his
son. There every year a famous festival is stiU held in summer
time by the people of the neighbourhood as well as by the inhabitants
of the more distant parts of Palestine and by the Phoenicians and
Arabians, Very many also assemble for trade, to buy and sell ; for
every one sets great store on the festival* The Jews do so because
they pride themselves on Abraham as their founder ; the Greeks do
so on account of the visit of the angels ; and the Christians do so
also because there appeared at that time to the pious man One who
in after ages made iiimself manifest through the Virgin for the
salvation of mankind. Each, after the manner of his faith, does
honour to the place, some praying to the God of all, some invoking
the angels and pouring wine, or offering incense, or an ox, or a goat,
or a sheep, or a cock. For every man fattened a valuable animal
throughout the year, vowing to keep it for himself and his family to
feast upon at the festival on the spot And all of them here refrain
from women, either out of respect to the place or lest some evil
should befall them through the wrath of God, though the women
beautify and adorn their persons specially, as at a festival, and show
themselves freely in public. Yet there is no lewd conduct, though
the sexes camp together and sleep promiscuously. For the ground
is ploughed and open to the sky, and there are no houses except the
ancient house of Abraham at the oak and the well that was made by
him. But at the time of the festival no one di^aws water from the
well* For, after the Greek fashion, some set burning lamps there ;
others poured wine on it, or threw in cakes, money, perfumes, or
incense. On that account, probably, the water was rendered unfit
to drink by being mixed with the things thrown into it The
performance of these ceremonies according to Greek ritual was
reported to the Emperor Constantine by his wife's mother, wlio had
gone to the place in fulfilment of a vow/ ^

' Sozomenus, liistoHa Ecdesiiistkaf ii 4 (Migne's Fairologia Grmca^ btvii.
941, 944}.

FOLK-LORE IN THE OLD TESTAMENT

131

Thus it appears that at Hebron an old heathen worship of the •
sacred tree and the sacred weO sui*vived in full force down to the
estabUshment of Christianity, Tlie fair wliich was held along with
the summer festival appeal's to have drawn merchants together from
many quarters of the Semitic world. It played a melancholy part '
in the history of the Jews ; for at this fair, after the last siege and
destruction of Jerusalem by the Romans under Hadrian in the
year 119 a. b,, a vast multitude of captive men, women, and childi*en
was sold into slavery.^ So the Jewish nation came to an end on the •
very spot where it was traditionally said to have been founded by
Abraham, at the sacred oak or terebinth of Mamre. The tree, or
rather its successor, is shown to this day in a grassy field a mile and
a half to the west of Hebron. It is a fine old evergreen oak {Quercm
pseudo-coccifera), the noblest tree in southern Palestine. The trunk
is twenty-three feet in girth, and the span of its spreading branches
measures ninety feet. Thus in the long rivalry between the oak
and the terebinth for the place of honour at Mamre the oak has won.
There is not a single large terebinth in the neighbourhood of Hebron.^

§ B. The Covenant on the Cairn.

When Jacob fled from Paddan-aram with hLs wives and his
children, his camels and his cattle, Laban pursued after him and
came up with the long lumbering train of fugitives in the be^utifiil
wooded mountains of Gilead, to the east of the Jordan. The two
kinsmen agreed to make a covenant, and for that purpose they
gathered stones, piled them up into a cairn to be a witness between
them, and pai*took of food on the cairn. ^ Here the eating of food

^ Jerome, Oommentaty on Jeremiah^ xxxi (Migne's Patrologia Latino^ xxiv. 877) ;
Cliwnicon Faschak, ed. L. Dmdorf; i. 474.

' Edward Kobinson, BMical Besearchm in Palestine^ ^ ii 81 sq. ; W. M* Thomson,
The Ijand and the Booky SoufJiem Pakstine and Jerusalem^ pp. 282-4 ; H, B. Tristram,
'me Land of Israel \ pp. 382-4 j id.. The Naiural Histofy of the Bible \ p, 369;
Baedeker, Palestine and Si/ria\ p, 115*

' Genesis xxxi, 17-55. In verse 46 the Revised Version ti^anslates : * and they
did eat there by the heap/ where the Authorized Version renders: *and they did
eat there upon the heap/ The parallels which I adduce in the text make it prob-
able that the Authoiized Version is here right and the Revised Version wrong.
The primary sense of the preposition in question (/?) is certainly ' upon ', and there
is no reason to depart from it in the present passage.

K 2

132

FOLK-LORE IN THE OLD TESTAMENT

upon the stories was probably intended to ratify the covenant.
How it was supposed to do m may perhaps be gathered from a Norse
custom described by the old Danish historian Saxo Grammaticus.
He t^lls us that * the ancients, wlien they were to choose a king,
were wont to stand on stones planted in the ground, and to proclaim
their votes, in order to foreshadow from the steadfastness of the
stones that the deed would be lasting/ ^ In reaUty the stability of
the stones may have been thought to pass into the person who

r stood on them and so to confirm his vote. Thus we read of
a mythical Rajah of Java who bore the title of Kajah Sela Perwata,
'which in the common language is the same as Watu Gunung,
a name conferred upon him from his having rested on a mountain
like a stone, and obtained his strength and power thereby, without

•'other aid or assistance.' - At a Brahman marriage in southern
India * the bridegroom takes up in his hands the right foot of the
bride, and places it on a mill-stone seven times. This is known as
swptupadi (seven feet), and is the essential and binding portion of
the marriage ceremony. The bride is exhorted to be as fixed in
constancy as the stone on wliich her foot has thus been placed.' *
Similarly at initiation a Brahman boy is made to tread with his
right foot on a stone, while the words are repeated : * Tread on this
stone; like a stone be firm/^ Among the Kookies of Northern
Cachar at marriage ^ the young couple place a foot each upon a lai^e
stone in the centre of the village, and the GhaHm [head-man]
sprinkles them with water, and pronounces an exhortation to general
virtue and conjugal fidelity; togetlier with a bleasing and the ex-
pression of hopes regarding numerous progeny','"' In Madagascar
it is believed tliat you can guard against the instabihty of earthly
bliss by burying a stone under the main post or under the threshold

* Tim First Nine Books o/fJiC Danish Historf/ ofSojro Gmmmaiicus, translated by
0. Elton, p. 16. The original runs thus : Lcdim regem vekrcs affixis Mono saxis
imistetr suffragiaque jyromere conmicverant, suhjectof-um lapidum firtmtaie fatii ominaturi
(Hi^oria Banica, lib. i., p. 22, ed. P. E. Mailer).

* T, S. Baffles, Ilistori/ of Jam (London. 1817), i, 377.

' E. Thurston^ Ethnofjfraphic Notes in tSotitheni India (Madras, 1W6), p. 1.
Compare Sonnerat, Vmjage mix Indcs Oriat talcs et a la Chine (Paris, 1782), i. 8L

* Griht/Q'Siitras, translated by H. Oldenberg, Part ii., p. 146 (Sacred Books of
the likist, vol xxx),

* Lieut< R. Stewart, * Notes on Noiihern Cachar,* Journal of tlw AsiaUc Socichf
ofBef}ffal xxiv (1855), pp. 620 sq.

FOLK-LORE IN THE OLD TESTAMENT

138

of your house, 1 The weight of the stone is clearly supposed to
counteract the levity of fortune.

On tlie same principle we can explain the custom of swearing •
with one toot or with both feet planted on a stone. The idea seems
to be that the solid enduring quality of the stone will somehow pass
into the swearer and so ensure that the oath will be kept. Thus
there was a stone at Athens on which the nine archons stood when
they swore to rule justly and according to the laws.^ A little to
the west of St. Colnmba's tomb in lona * lie the black stones, wMch
are so called, not from their colour, for that is grey, but from the
effects that tradition says ensued upon perjury^ if any one became
guilty of it after swearing on these stones in the usual manner;
for an oath made on them was decisive in all controvei'sies.
Mae-Donald, king of the isles, delivered the rights of their lands
to his vassals in the isles and continent, with uplifted hands and
bended knees, on the black stones ; and in this posture, before
many witnesses, he solemnly swore that he would never recall
those rights which he then granted ; and this was instead of his
great seal. Hence it is that when one was certain of what he ^
affirmed he said positively, I have freedom to swear this matter
upon the black stones/ ^ Again, in the island of Fladda, another %
of the Hebi*ides, there used to be a round blue stone on wliich
people swore decisive oaths.^ When two Bogos of eastern Africa
have a dispute, they will sometimes settle it at a certain stone,
which one of them mounts. His adversary calls down the most
dreadftil eui*ses on him if he forswears liimself, and to every cui'so
the man on the stone answers * Amen ! ' ^* At Ghosegong in the
Garrow Hills of north-eastern Bengal, thei'e is a stone on which
the natives swear their most solemn oaths. In doing so they first
salute the stone, then with their hands joined and uplifted, and
their eyes steadfastly fixed on the hUls, they call on Maliadeva to

(

' Father Abiimle, * Afitrologie Mftlgaehe/ Missions CathoUques^ xi (1879). p. 482;
Qui va mttrr^r an caiUmi mi pied du grand potcau dc la case oti sous k seuil de la
porttf a Veffet de se donfi^r un dcstin de poids et de JulHite, aprds satire lav^ Sun destin
ifinconsfance,

^ Aristotle, Constitution of Athens, 7 and 55 ; Plutarch, Solon^ 26 j Pollux,
viii, 86,

^ Martin, * Description of the Western Islands of Scotland/ in Pinkerton'a
Voyages and Tmmis, iii- 657,

* Martin, op, ciL^ pp. 627 sq.

* W. Muniinger, Sitten und Hedd d^r Bogos (Winterthur, 1859), pp. 33 sq.

136

FOLK-LORE IN THE OLD TESTAMENT

had closed on him his name might never again be mentioned under
pain of death. There traitors rotted, and there the perjured king of
Armenia ended his days, '

§ 4. Jacob at the ford of the Jahhok.

After pai*ting from Laban at the cairn, Jacob, with his wives
and children, his flocks and his herds, pui-sued his way southward
to meet his brother Esau. From the breezy, wooded heights of the
mountains of Gilead he now plunged down into the profound ravine
of the Jabbok thousands of feet below. The descent occupies several
hours, and the traveller who aecomplishes it feels that, in reaching
the bottom of the deep glen, he has passed into a different climate.
From the pine-woods and chilly winds of the high uplands he
descends first in about an hour*s time to the balmy atmosphere of
the village of Burmeh, embowered in fruit-trees, shrubs, and flowers,
where the clear cold water of a fine fountain will slake his thirst at
the noonday rest. Still continuing the descent, he goas steeply down
another two thousand feet to find liimself breathing a hot^house aii-
amid luxuriant semi-tropical vegetation in the deptlis of the great
lyn of the Jabbok, The gorge is in the highest degree wild and
picturesque. On either hand the cliffs rise almost perpendicularly
to a great height ; you look up the precipices or steep banks to
the sky-line fai' above, At the bottom of this mighty chasm the
Jabbok flows with a powerful current, its blue-grey water fringed
and hidden even at a short distance by a dense jungle of tall
oleandei-s, whose crimson blossoms add a glow of colour to the
glen in early summer. The Blue River, for such is its modern
name, runs fast and strong. Even in ordinary times the water
reaches to the horses girths, and sometimes the stream is quite
unfordable, the flood washing grass and bushes high up the banks
on either hand. On the opposite or southern side the ascent from
the ford is again exceedmgly steep. Tlie path winds up and up ;
the traveller must dismount and lead his horse.- It waB up that

' PrcMjopius, Be Bella Persico, i, 5.

' W. M. Thomson, The Land and fJte Book, JxhanQn, BanmscHS, and betfOHd
Jordm, pp. 588 sqq. ; H. B. Tristram, The Land oflsraeP, p. 519. The ford her©
described is that of Mukhadat en Nusranlyeh, *the Ford of the Christbii Woman/ on
the road between Beimftn and Shihdn. It is the ford on the regular road from north
to south, and is probably therefore the one at which tradition plat:ed the passage

FOLK-LORE IN THE OLD TESTAMENT

137

long ascent that Jacob, lingering alone by the ford in the gloaming,
watched the camels labouring and heard the cries of the drivers
growing fainter and fainter above liim, till sight and sound of them
alike were lost in the darkness and the distance.

The scene may help us to understand the strange adventure
which befell Jacob at the passage of the river. He had sent his
wives, his handmaids, and his children, riding on camels, across the
river, and aD his flocks and herds had preceded or followed them.
So he remained alone at the ford. It was night, probably a moon* •
light summer night ; for it is unlikely that with such a long train he
would have attempted to ford the river in the dark or in the winter,
when the current would ran fast and deep. Be that as it may,
in the moonlight or in the dark^ beside the rushing river, a man •
wrestled with liim all night long, till morning flushed the wooded
crests of the ravine high above the struggling pair in the shadows
below. The stranger looked up and saw the light and said, * Let me ♦
go, for the day breaketh/ So tlie ghost of Hamlet's father faded at
cockcrow; so Mephistopheles in the prison warned Faust, with
the hammering of the gallows in his ears, to huny, for the day,
Gretchen*s last day, was breaking. But Jacob clung to the man »
and said, * I will not let thee go, except thou bless me.' The *
stranger asked him his name, and when Jacob told it he said,
'Tliy name shall be called no more Jacob, but Israel; for thou
hast striven with God and with men, and hast prevailed.' But »
when Jacob inquired of him, *Tell me, I pray thee, thy name/ the
man i^efused to mention it, and having given the blessing which
Jacob had extorted, he vanished. So Jacob called the name of the *
place Peniel, that is the Face of God ; ' For/ said he, * I have seen
God face to face, and my life is preserved.* Soon filter the sun •
rose and shone on Jacob, and as it did so he limped ; for in the
struggle his advemary had touched liim on the hollow of the thigh.

of Jacob with hia family and his herds. In degcribing the gorge and the ford
I have followed closely the accounts of Thomson and Tristram, who both passed
tlm way and wrote as eye-witnesses, A very different impression of the scenery
of the Jabbok is given by Professor G. A. Smith s eloquent description (Hisioricul
Geography of the Hohj Laml, p. 684), which probably applies mainly either to the
upper or the lower reaches of the river, before it has entered the great caiion^ or
after it has emerged from it into the broad strath of the Jordan. In these districts,
accordingly, it would seem that the aspect of the river and its banks is one of
pastoral peace and sweet rural charm, a landscape of Constable rather than of
Salvator Bosa.

138

FOLK^LOEE IN THE OLD TESTAMENT

* Therefore the children of Israel eat not the sinew of the hip which
is upon the hollow of the thigh unto this day : because he touched
the hollow of Jacob s tliigh in the sinew of the hip/ ^

The story is obscure, and it is probable that some of its original
features have been deliberately modified or omitted by the compilers
of Genesis because they savoured of heathendom- Hence any
explanation of it must be to a great extent conjecturaL But
taking it in connexion with the natural features of the place
where the scene of tlie story is laid, and with the other legends
of a similar character wliich I shall adduce, we may perhaps
provisionally suppose that Jacob s mysterious adversary was the
spirit or jinnee of the river, and that the struggle was purposely
sought by Jacob for the sake of obtaining his blessing. Tliis would
explain why he sent on his long train of women, servants, and
animals, and waited alone in the darkness by the ford. He might
calculate that the shy river-god, scared by the trampling and
splashing of so great a caravan through the water, would lurk in
a deep pool or a brake of oleanders at a safe distance, and tliat when
all had passed and silence again reigned, except for the usual
monotonous swish of the current, curiosity would lead him to
venture out from his lair and inspect the ford, the scene of all
this hubbub and disturbance. Then the subtle Jacob, lying in wait,
would pounce out and grapple with him till he had obtained the
coveted blessing. It was thus that Menelaus caught the shy sea-god
Proteus sleeping at high noon among the seals on the yellow sands,
and compelled him reluctantly to say his sooth. ^ It was thus that
Peleus caught the sea-goddess Thetis and won her, a Grecian Undine,
for his wife.^ In both tliese Greek legends the supple, slippery
water-spirit writhes in the grip of his or lier captor, slipping
through his hands again and again and shifting his or her shape
fix)m lion to serpent, from serpent to water, and so forth, in the
effort to escape j not till he is at the end of all his sliifts and sees
no hope of evading his detemiined adversary does he at last consent
to grant the wLshed-for boon. So, too, when Hercules wrestled
with the river-god Achelous for the possession of the fair Dejanira,
the water-sprite turned himself first into a serpent and then into

' Genesis xxxl Si-xxxii. For the camels on which Jacob's family rode, see
id, xxxi. 17,

■ Homer, Odijssmf iv. 354-570,

' ApoUodorus, iii, 18. 5 \ Scholiast on Pindar, Ncm. iii. 60,

FOLK-LORE IN THE OLD TESTAMENT

139

a bull in order to give the brawny hero the slip; but all in
vain.^

These parallels suggest that in the original form of the tale
Jacobs adversary may in like manner have shifted his shape to
evade his importunate suitor. A trace of such divine metamorphoses
perhaps survives in the story of God's revelation of himself to Elijah
on Mount Horeb ; the wind, the earthquake, and the fire in that
sublime nanative may in the first version of it have been disguises
assumed, one after the other, by the reluctant deity until, vanquished
by the prophet s perseverance, he revealed himself in a still small
voice. ^ For it is to be observed that water-spirits are not the only
class of supernatural beings for whom men have laid wait in order
to wring from them a blessing or an oracle. Thus the Phrygian
god Silenus is said, m spite of his dissipated liabits, to have possessed
a large stock of general information wliich, Hko Proteus, he only
imparted on compulsion. So Midas, king of Phrygia, caught him
by mixing ^vine with the water of a spring from which, in a moment
of weakness, the sage had condescended to drink. When he woke
from his drunken nap^ Silenus found himself a prisoner, and he had
to hold high discoui-se on the world and the vanity of human hfe
before the king would let him go. Some of the gravest writers of
antiquity have bequeathed to us a more or less accurate report
of the sermon which the jolly toper preached beside the plasliing
wayside spring or, according to others, in a bower of roses. ^ By
a stratagem like that of Midas it is said that Numa caught the rustic
deities Picus and Faunus and compelled them to draw down Jupiter
himself from the sky by their charms and spells. ^

The view that Jacob s adversary at the ford of the Jabbok was
the river-god himself may perhaps be confirmed by the obsei'vation
that it has been a common practice with many peoples to propitiate
the fickle and dangerous spirits of the water at fords. Hesiod says
that when you are about to ford a river you should look at the
running water and pray and wash your hands ; for he who wades

* Ovid, Metamorph, ix. 6^86; compare Sophocles, Tmchiniaef 9-21.
■ 1 Kings xii. 8-13.

* Xenophon, AnabasiSj i. 2. 13 ; Pausanias, i. 4. 5 ; Herodotus, viii. 188 ;
Plutarcli, Cofisol ad ApoJlan, 27 ; Aelian^ Var, Hist iii, 18 ; Pliiloatratus, ViL
ApoUon, vi. 27; Himerius, Ecloff, xvi. 5; Cicei*o, TuscuL DispuL u 48, 114;
Servius on Virgil, Eel vl 13.

* Ovid, Fasti, iii. 289-348.

140

FOLK-LORE IN THE OLD TESTAMENT

through a stream with unwashed hands mcurs the wrath of the
gods J When the Spartan king Cleomeiies, intending to invade
Argolis, came with his array to the banks of the Erasiiius, he
sacrificed to the river, but the omens were unfavourable to his
crossing. Thereupon the king remarked that he admired the
patriotism of the river-god in not betraying his people, but that
he would invade Argolis in Hpito of him. With that he led his
men to the sea-shore, sacrificed a bull to the sea, and transported
his army in ships to the enemy's country,' When the Persian host
under Xerxes came to the river Strymon in Thrace, the Magians
sacrificed white horses and perfonned other strange ceremonies
before they crossed the stream.^* Luculhis at the head of a Koman
army sacrificed a bull to the Exiphrates at his passage of the river. ^

* On the river-bank the Peruvians woidd scoop up a handful of water
and drink it, praying the river-deity to let them cross or to give
them fish, and they threw maize into the stream as a propitiatory
offering ; even to this day the Indians of the Cordilleras perform the
ceremonial sip before they \vill paas a river on foot or horseback/ ^

* It is a custom among native tribes of South Africa to pay respect
to rivers, which would appear to intimate that formerly they were
woi^shippedj or rather that individual rivers were supposed to be
the dwelling-place of a spirit. Thus, when a river has been safely
crossed, it is the custom in some parts to throw a stone into its
watei*s, and to praise the itom/o, . . . When Dingan's army was going
against Umzilikazi, on reaching the banks of the Ubulinganto, they
saluted it, sayingj ** Sa ku honUy huHnganto^'' and having strewed animal
charcoal {umsizi) on the water, the soldiers were made to drink it.
The object of this was to deprecate some evil power destructive to
life, which was supposed to be possessed by the river.'' From
another writer we learn that Caffies spit on the stones which
they throw into the water at crossing a river. He tells us that
formerly they * were in the habit of either sacrificing some animal

* Hesiod, Works and Ikiffs, 737-41* As to the Greek worship cjf rivers, see the
evidence collected by R. Karsten, Shulies in primitive Qreck religion (Helsingfors,
1907), pp. 29&(iq.

* Herodotus^ vi 76.

* Herodotus* vil 113,

* Plutarch, LttmUus, 24.

* E. B. Tylar, Primitive Culture^ il. 210.

■ Callaway, Nursertf TaleSy Traditions and Histories of the Zuhts, u 90, not© -^.

FOLK-LORE IN THE OLD TESTAMENT

or oflfering some grain to appease ancestral spirits living in the river,
Tlie Bushmen used to offer up some game they had killed, or in the
absence of that would offer up an ai-mw/ ^ A tliird writer informs
lis that in the belief of the Bantu tribes of south-eastern Africa ' rivers
are inhabited by demons or malignant spirits, and it is neceasary
to propitiate these on crossing an unknown stream, by throwing a
handful of corn or some otlier offering, even if it is of no intrinsic
value, into the water ',- When the Masai cross a stream they throw
a handful of grass into the water as an offering ; for gi*ass, the source
of life to their cattle, plays an imjwrtant part in Masai superstition
and ritual.^

Tlie Badagas, a tribe of the Neilgherry Hills m southern India,
believe in a deity called Gangamma, ' who is supposed to be present
at every stream, and esi>ecially so at the Koonde and Pykare rivers,
into which it was formerly the practice for every owner of cattle,
which had to cross them at their height, to tlnow a quainter of
a rupee, becauye their cattle used frequently to be carried away by
the current and destroyed. It is enumerated amongst the greatest
Bins of every deceased Badaga at Iiis funeral that he had crossed
a stream without paymg due adoration to Gangamma/^ Again, the
Todas, another smaller but better known tribe of the same hills^
regard two of their rivei^s, the Teipakh {Paikara) and the Pakliwar
(Avalanche), as gods or the abodes of gods. Every person in crossing
one of these streams must put his right arm outside of his eloalc in
token of respect. Formerly these; rivei^ miglit only be crossed on
certain days of the week. When two men who are sons of a brother
and a sister respectively pass in company over either of the sacred
streams they have to perform a special ceremony. As they approach
the river they pluck and chew some grass, and each man says to the
other, * Shall I throw the river (water) ? Shidl I cross the river?'
Then they go down to the bank, and each man dips his hand in the
river and throws a handful of water away from him thrice. After
that they cross the river, each of them with his arm outside of his
cloak in the usual way. But if the day is a Tuesday, Friday, or
Saturday they will not throw tlic water, but only chew the grass.

^ Dudley Kidd, TIte Esamiial K(i/ir (homlon. 1904), p. 10.
* J. Macdonald, LiglU in A^Mca'^ (London, 1890K p. '*205. Compare id,, in
Journal of the Aftthropohsfical Imtiiule, xx (1891), p. 125.

= a L. and H. Hinde, Tlie Last of tlie 3lasm (London, 190 IK pp. 103 sq.
^ F. Metz, The Tribes of the Neilgherru Hills (Mangalore, 1864), p, 68.

142

FOLK-LORE IN THE OLD TESTAMENT

Also, if the funeral ceremonies of a person belonging to the clan of
either of the two men are not complete, they will not throw the
water. The sacred dairyman {palol) of the Todas may not cross
either of the holy rivers at the places used by common folk. In the
old days there were certain fords where ordinary people waded
through the water, but the dairyman had a ford of his own^
Nowadays the Todas cross the Paikara by a bridge, but the holy
milkman may not make use of the profane convenience. And in
the old days no Toda who had been bitten by a snake might cross
any stream whatever,^ Among the Mahafaly and Sakalava of
Bouthern Madagascar certain chiefs ai*e forbidden to cross certain
rivers, while others are bound to go and salute all the rivers of the
country. 2 In Cayor, a district of Senegal, it is believed that the
king would inevitably die within the year if he were to cross a river
or an arm of the sea.^

Though we may not be able to explain the exact reasons for
imposing these various rules and restrictions, the general motive
which underlies them is plain enough; it is the awe and fear of
rivers conceived as powerful personal beings. That conception is
well illustrated by a practice observed by the Kakhyeen of Upper
Burma. When one of the tribe has been drowned in crossing a
river, the avenger of blood repairs once a year to the banks of the
guilty stream, and filling a vessel full of water he hews it through
with his sword, as if he were despatching a human foe.* The same
tendency to personify the spirit of a river, especially a rapid and
dangerous river, perhaps explains the weird stoi-y of Jacob's adven-
ture at the ford of the Jabbok.

Tlie tradition that a certain sinew in Jacob's thigh was strained
in the struggle with his nocturnal adversary is clearly an attempt
to explain why the Hebrews would not eat that particular sinew.
Both the tradition and the custom have their parallels among some
tribes of North American Indians, who regidarly cut out and throw
away the hamstrings of the deer they kilL Without repeating
the evidence on this subject which I have cited elsewhei'e, • I will

' W. H, R. Rivers, The Todas, pp, 418 sq., 500 sq.

■ A. Van Gennep, 'Ttiboii et Tot^misme a Madagascar {F&riB, 1901)^ p, 118,

* J, B. L, Durand, Vmjagt an SdfukffM (Paris, 1802), p. 55.

* element Williams, Through Burma to Western CMtia (Edinburgh and London,
1868), pp. 91 sq.

' TIic Gdikn Bough\ \u 419-21.

FOLK-LORE IN THE OLD TESTAMENT 143

merely mention two reasons which the Cherokee Indians assign
for the practice. One is that * this tendon, when severed, draws
up into the flesh ; ergo, any one who should unfortunately partake
of the hamstring would find his limbs draw up in the same
manner/^ The other is that if they did not do so they would
easily grow tired in travelMng.- Both these reasons assume tiie
principle of sympathetic magic, though they apply it differently.
The one supposes that, if you eat a sinew which shrinks, the cor-
responding sinew in your own body will shrink likewise, Tlie
other appears to suppose that, if you destroy the sinew without
which the deer cannot walk, you yourself will be incapacitated
from walking in precisely the same way* Both reasons are
thoroughly in keeping with savage philosophy. Either of them
would suffice to account for the Hebrew taboo.

§ 5. The Bundle of Life.

When David with his men was in hiding for fear of Saul in
the dreary wilderness of Judaea^, he was visited by Abigail, the
wise and beautiful wife of the rich sheep-farmer Nabal, whom the
gallant outlaw had laid under a deep obligation by not stealing
his sheep* Insensible of the services thus rendered to him by the
cat^rans, the surly farmer refused with contumely a request, couched
in the most poUte terms, which the captain of the hand had sent
in for the loan of provisions* The msult touched the captain's
nice sense of honour to the quick, and he was marchmg over the
hills at the head of four hundred pretty fellows, every man of
them with his broadsword buckled at his side, and was making

' J* Mooney, * Sacred formulas of the Cherokees,' Seventh Annual Report of the
Bureau of Ethnology (Washington, 1891), p* 823.

• J. Mooney, ' Myths of the Cherokees/ Nineteenth Annual Eeport of the Bureau
of American EtJimloffff (Washington, 1900), p. 263.

' Speaking of the wildernesa of Judaea, an eye-witness says : * The view from
the height was most extraordinary ; on every side were other ridgea equally white,
steep, and narrow; their aides were seamed by innumerable torrent-heds, their
summits were sharp and ragged in outline. These ndges stood almost isolated,
between broad flat valleys of soft whit© marl scattered with filnts, and with a
pebbly torrent-course in the middle. There was not a tree visible, scarcely even a
thorny shrub ; the whole was like the dry basin of a former sea scoured by th«^
rains, and washed down in places to the hard foundation of metamoi^pliic limestone,
which underlies the whole district, and forms precipices two thousand feet high
over the shores of the Dead Sea.' (C. R, Conder, Tent-work inFalestine, ii. 127*)

FOLK-LORE IN THE OLD TESTAMENT

straight for the farm, wlien the farmers wife met Iiini on the
moor. She had soft words to soothe the ruffled pride of the angry
chieftain, and, better perhaps than words, a train of asses laden
with meat and drink for the sharp-set brigands. David was melted.
The beauty of the woman, her gentle words, and the sight of the
asses with their panniers, all had their effect. He received the
wife, pleading for her husband, with the utmost courtesy, promised
liis protection, not without dai'k hints of the sight that the sun
would have seen at the farm next morning if she had not met
liim, and so dismissed her with a blessing. The word was given.
The outlaws faced to the right-about, and, followed no doubt by
the asses, marched off the way they had come. With a hghter
heart Abigail hastened to the house whei*e her boorish husband
and his hinds, Uttle wotting of what had passed on the hills, were
drinking deep and late after the sheep-shearing. That night over
the wine she wisely said nothing. But next morning, when he was
sober, she told liim, and his heart died within liim. The shock
to his nervous system, or perhaps sometliing stronger, was too
much for liim. Within ten days he was a dead man, and after
a decent interval the widow was over the hills and far away with
the captain of the brigands.^

Among the compliments wliich the chamiing Abigail paid to

. the susceptible David at their first meetmg there is one which
desei-ves our attention. She said : * And though man be risen up
to pui-sue thee, and to seek thy soul, yet the soul of my lord sliall
be bound in the bundle of li fe with the Lord thy God j and the
souls of thine enemies, them shall he sHng out, as from the hollow
of a sling/ - No doubt the laiiguage is metaphorical, but to an

' EngUsh reader the metaphor is strange and obscure. It implies
that the souls of living people could be tied up for safety in a
bundle, and that on the contrary, when the souls were those of
enemies, the bundle might be undone and the souls scattered to
the winds, I think we may safely say that such an idea could
hardly have occurred to a Hebrew even as a figure of speech, unless
he were familiar with an actual belief that souls could thus be

* 1 Samuel xxv. 1-37.

^ 1 Samuel xxv. 29, The same expression * bundle of life ' {Q^*n inx) is
applied to a faithful friend in the Hebrew text of Ecclesiastic us, vi, 16, wher^
nin^ (^ bundle ') ought not, with some editors, to be changed into ^1* (* balm '}, See
Professor A. A. Bevan, in Jounial of Theological StudicSf October, 1899, p, 140,

FOLK-LORE IN THE OLD TESTAMENT

treated* To us who conceive of a soul as immanent in its body
so long as life lasts the idea conveyed by the verse in question is
naturally preposterous. But it would not be so to many peoples
whose theory of Ufe differs widely from om-s. There is in fact
a widespread belief among savages that the soul can be, and often
is, extracted from the body during the lifetime of its owner without
immediately causing his de^ith. Commonly this is done by ghosts,
demons, or evil-disposed persons who have a grudge at a man and
steal his soul for the pm-pose of killing him ; for if they succeed
in their fell intent and detain the truant soul long enough the man
will fall ill and die.^ For that reason people who identify their
souls with their shades or reflections are often in mortal terror
of a camera, because they think that the photographer who has
taken their likeness has abstracted their souls or shades along
with it. To take a single instance out of a multitude. At a village
on the lower Yukon River in Alaska an explorer had set up his camera
to get a picture of the Esquimaux as they were moving about among
their houses. While he was focussing the instrument, the headman
of the village came up and insisted on peeping under the cloth.
Being allowed to do so he gazed agog for a minute at the moving
figures on the ground-glass ; then jerking his head from under
the cloth he bellowed out to his people, ' He has got all your
shades in this box/ A panic ensued among the group, and in
a twinkling they disappeared helter-skelter into their houses.^ On
this theory a camera or a packet of photographs is a box or bundle
of souls packed ready for transport like sardines in a tin.

But sometimes souls are extracted from their bodies with a
kindly intention. The savage seems to think that nobody can die
properly unless his soul is in his body just before he expires, since
it is the final departure of the soul which is the true cause of death.
From this again he infers that if you can only draw out the soul
and keep it in safe custody the man in the meantime is for aU
practical purposes immortal, since in the absence of his soul there
is realty nothing in liini to die. Hence in time of danger the wary
savage will sometimes carefully extract his own soul or the soul
of a friend and leave it, so to say, at deposit account in some safe

* TM OiMen Bough ^, ii. 263 sqq* ; A, C. Kniijt^ Het onimiamB in den Indischen
Archipel {The Hague, 1906), pp. 77sc]q.

' E. W. Nelson, * The Eskimo about Behring Strait,* Eighteenth Anntml Report
qf the Bureau of Afnerican Ethnology^ part i (Washington, 1899), p. 422.

146

FOLK-LORE IN THE OLD TESTAMENT

place tin the danger is past and he can reclaim liis spiritual property*
For example^ many people regard the removal to a new house as
a crisis fraught with peril to their souls ; hence in Minahassa,
a district of Celebes, at such critical times the priest collects the
souls of the whole family in a bag and keeps them there till the
danger is over, when he restores them to their respective owners^*
Again, in Southern Celebes, when a woman's time is near, the
messenger who goes to fetch the doctor or midwife takes with him
a chopping-knife or something else made of iron. The thing, what-
ever it is, represents the woman's soul, which at this dangerous
time is believed to be safer outside of her body than in it. Hence
the doctor must take gmat care of the thing, for wem it lost the
woman's soul would with it be lost also. So he keeps it in his
house till the confinement is over, when he gives back the precious
object in return for a fee.^ In the Key Islands a hollowed-out
cocoa-nut, spht in two and carefully pieced together, may sometimes
be seen hanging up. Tliis is a receptacle in which the soul of
a newly-born infant is kept lest it should fall a prey to demons.
For in those parts the soul does not permanently lodge in its
tabernacle of clay, until the clay has taken a firm consistency.
The Esquimaux of Alaska adopt a similar precaution for the soul
of a sick child. The medicine-man conjures it into an amulet
and then stows the amulet in his medicine-bag, where, if anywhere,
the soul should be out of hai'm's way,^

But perhaps the closest analogy to the * bundle of life ' is
furnished by the bundles of churitiga^ that is, flattened and elongated
stones and sticks, which the Arunta and other tribes of Central Aus-
traha keep with the greatest care and secrecy in caves and crevices of
the rocks. Each of these mysterious stones or sticks is intimately
associated with the spii'it of a member of the clan, hving or dead ;
for as soon as the spirit of a child enters into a woman to be bom,
one of these holy sticks or stones is dropped on the spot where
the mother felt her womb quickened. Directed by her, the father

* P, N, Wilken, ' Bijdragen tot de kennis van de zeden en gewoonten der
Alfo€ren in de Minahassa.' Mededeelin^en van tmge hei Nederkindsche Zendelm^
genootschup, vii {1863). pp. 146 sq.

* B» F. Matthes, BUdragen tot de Ethndogie mn Zuid-Cehbes (The Hague, 1875),
p. 54.

' J. A. Jacobsen, Eeisen in die Insclwdt des Banda-Meeres (Berlin, 1896)»
p. 199.

POLK.LORE IN THE OLD TESTAMENT

searches for the stick or stone of his child, and having found it, or
carved it out of the nearest hard-wood tree, he delivei-s it to the
headman of the district, who deposits it with the rest in the sacred
store-house among the rocks. Tliese precious sticks and stones,
closely bound up with the spirits of all the members of the clan,
are often carefuUy tied up in bundles. They constitute the most
sacred possession of the tribe, and the places where they are deposited
are skilfully screened from observation, the entrances to the caves
being blocked up with stones arranged so naturally as to disarm
suspicion. Not only the spot itself but its surroundings are sacred. *
The plants and trees that grow there are never touched : the wild
animals that find their way thither are never molested. And if a
man fleeing from his enemies or from the avenger of blood suc-
ceeds in reaching the sanctuary, he is safe so long as he remains
within its bounds. The loss of their churifiga^ as they call the sacred
sticks and stones thus associated with the spirits of all the living
and all the dead members of the community, is the most serious evil
that can befall a tribe. Robbed of them by inconsiderate white men,
the natives have been known to stay in camp for a fortnight, weeping
and wailing over their loss and plastering their bodies with white
pipeclay, the emblem of mourning for the dead.^

In these beliefs and practices of the Central Australians with
regard to the churinga we have, as Messrs, Spencer and Gillen justly
observe, *a modification of the idea which finds expression in the
folklore of so many peoples, and according to which primitive man,
regarding his soul as a concrete object, imagines that he can place it
in some secure spot apart, if needs be, from his body, and thus, if
the latter be in any way destroyed, the spirit part of him still persists
unharmed* ^ Not that the Arunta of the present day beheve these
sacred sticks and stones to be the actual receptacles of their spirits
in the sense that the destruction of one of the sticLs or stones woidd
of necessity involve the destruction of the man, woman, or clxild
whose spirit is associated with it. But in their traditions we meet
with clear traces of a belief that their ancestors did really deposit
their spirits in these sacred objects. For example, we are told that
some men of the Wild Cat totem kept their spirits in their churinga^
which they used to hang up on a sacred pole in the camp when they

* Spencer and Gillen, The Naiive Dribes of Central AuatmUa, pp» 128-56.
Compare lU, The Nor^em Tribes of Central AustraliUy pp. 267-82.

* Spencer and Gillea, Native Tribes of CefUral Australia^ p. 137.

L 2

FOLKLORE IN THE OLD TESTAMENT

went out to hunt ; and on their retimi from the chase they would
take down the churinga from the pole and carry them about as
before,^ The intention of thus hanging up the churittga on a pole
when they went out hunting may have been to put their souls in
safe-keeping till they came back.

Thus there is fair ground to think that the bundles of sacred
sticks and stones, which are still treasured so carefully in secret
places by the Arunta and other tribes of Central Austraha, were
formerly beheved to house the souls of every member of the
community. So long as these bundles remained securely tied up
in the sanctuary, so long, might it be thought, was it well with the
souls of all the people ; but once open tlie bundles imd scatter their
precious contents to the winds, and the most fatal consequences
would follow. It would be rash to assert that the primitive Semites
ever kept their souls for safety in sticks and stones which they
deposited in caves and crannies of their native wilderness ; but it is
not rash to affirm that some such practice would explain in an easy
and natural way the words of Abigail to the hunted outlaw : * And
though man be risen up to pursue thee, and to seek thy soul, yet the
soul of my lord shall be bound in the bundle of life with the Lord
thy God ; and the souls of thine enemies, them shall he sling out, as
from the hollow of a sling/

Thus I infer that the Hebrews retained down to historical times
the conception of an external soul, that is, a beHef in the possibility
of depositing the soul for safety in some secure place outside of the
body. The inference is confirmed by a remarkable expression of
Isaiah. In a long list of feminine ornaments he mentions * houses of
the soul ',* The expression thus Uterally translated is unique in the
Old Testament. Modern translators and commentatoi-s, following
Jerome, render it ^ perfume boxes % ' scent-bottles ', or the like.^ But
it may well be that these * houses of the soul * were amulets in which
the soul of the wearer was supposed to lodge,* The commentators on

' Spencer and Gillen, op* ciL^ p» 138.

* Isaiah iii. 20 C*B|n ^m.

* * Perfume boxes * (Revised Version). Similarly Ksutsch, Dillmann, Duiim,
Skiniier, Whitehouse. Jerome's rendering in the VuJgate ia dfactoriola,

* The Egyptians placed little models of houses, made of pottery, on the tombs
for the souls of the dead to lodge in. Many of these miniature houses of the soul
have lately been discovered by Professor W. M. Flinders Petrie at Kifeh, in Upper
Egypt See W. M. Flinders Petrie, Gizeh and Eifdi (London, 1907], pp. 14-20,
with plates I, XV-XXII* The hut-urns containing the ^hes of the dead which

FOLKLORE IN THE OLD TESTABfENT

149

the passage recogiuase that many of the trinkets in the prophet s list
were prahablj charms^ just as personal ornaments often are in the
East to the present day,* The very word which follows * houses of
the sool ^ in the text is rendered ^ amulets * in the Revised Versioii ;
it is derived from a verb meaning * to whisper \ * to charm *,*

But the view of these ^houses of the soul" which I have
suggested does not necessarily exclude their identification with
scent-bottles. In the eyes of a people who, like the Hebrews %
identified the principle of life with the breathy the mere act of
smelling a perfume might easily assume a spiritual aspect ; tlie
scented breath inhaled might seem an accession of life, an addition
made to the essence of the souL Hence it would be natural to regard
the fragrant object itself, whether a scent-bottle, inoensOi or a flower,
as a centre of radiant spiritual energy, and therefore as a fitting place
into which to breathe out the soul whenever for any reason it seemed
desirable to do so for a time. Far-fetched as this idea may appear to
us, it may seem natural enough to the folk and to their best inter^
preters the poets :

/ sent thee late a rosij wr&Ukf

Not so much hmwuring thee
As ffiting it a hope that there

It could not unlhep^i be;

haTe b€en found in aneieni Italian, German^ and Danish ^rmT«s, w«re probably
in li^e manner intended to aerre as houses of the soul. See W. Helbig, Dif
IlaiUsBr m <fer Poebene, p. 60 ; O, Schrader, Realkxikvn der Imlo^ermaniitkeH
AUffimmammde, pp. 337, 339.

> Dillmann, Skinner, and Whltehouae, on laaiah iii. 18 and 'KK Compare B.
Winer, BibliscMes BealitOtterbuch \ s, w * Amulete ', The peoples of the eastern horn
of Africa (the Somali* Gallas and Danakil), especially tlie Mohammedan part of
them, wear many ornaments which, at the same time, senre as amulets. See Ph,
Paulit^chke, Ethnographk Nordost-A/rikas^ Die matttieUe CSiUur dtr Dnmlkitt GaUa^
und Sontdl (Berlin, 1893), pp. 05 sq. Compare F. Stuhlmauo, Mit Emm I^Mscha m$
MefM von Afiika (Berlin, 1894)^ p, 518. On the relation of jewellery to niagic^
see Professor W* Ridgeway, in Iteport €^ Uie British Association, Meetit$if ktid at
Soufhport, 1903, pp. 815 sq.

' Brown^ Driver, and Briggs, Hebrctc ami English Lmcon qfthe Old Testaments
p, 588. Similarly Kautsch^ in his German translation of the Bible, and Oillniann
and Skinner in their commentaries on Isaiah. In auoiher passage (xxyj. I0|
Isaiah usea the same word (^u?} in the phr^e * compulsion of a spell * (where we
must read fiPt fer IV? with many critics; see Brown. Driver, and Briggs, op. ciL^
pp. 538, 848).

^ Genesis ii. 7.

160 FOLK-LORE IN THE OLD TESTAMENT

But thou thereofi didst mily l^eathe
And sent'st it back to me ;

Since tvhen it groivSj and smeUs^ I stvear^
Not qf itself ltd thee ! *

Or again :

Mr verhluhetj sUsse Hosen^
Meine Liebe trug euch nicht.

But if beauty can thus be thought to give of her life, her soul,
to the soul of the rose to keep it fadeless, it is not extravagant to
suppose that she can breathe her soul also into her scent-bottle. At
all events these old-world fancies^ if such indeed they are, would
explain very naturally why a scent-bottle should be called a * house
of the soul ', But the folk-lore of scents has yet to be studied. In
investigating it, as every other branch of folk-lore, the student may
learn much from the poets, who perceive by intuition what most of
us have to learn by a laborious collection of facts. Indeed without
some touch of poetic fancy it is hardly possible to enter into the
heart of the people. A frigid rationalist will knock in vain at the
magic rose-wreathed portal of fairyland. The porter will not open
to Mn Gradgrind.

^ * JoDBOn's learned sock ' was on when he wrote these beautiful verses. See

Philostratus, JEpisL 2 : Ilcn-o/i^a trot orci^at'Oi' poSiav^ ov <rf TtpuiK, teal rovTO fiivyap, aXX
avrai^ ri x^pt^op^vo^ toIs poBoi^, tva /a^ p^ctpavB^, And again E^ist. 46 : Ev TTCTTOojica?
OTput^vjj xpitjtrdfKvoi TOi? poSoi?" , . , et Sc ^ovXti rt ^iXw )(api^iG^at^ to. kti^ava avraiv
dvTiTTffuf/ov firjtitTi TTFCon^a p^mv pLOVOV^ aXXa Kal troiv* And the thought of the first

stanza of the same song:

Drink to me mtltf with thine eyes.

And I will pledge with mine;
Or Uave a kiss but in the cttp

And rU not look for winCj

is also borrowed from the same elegant writer* See Philostratus, EpiM, 33: 'Efioi
0€ povoifi wpoirtvt roLS opfUMfw • * . €£ Sc f3ovX.€tf rov p.lv olyov pij TrapaTroXAve, puovov
S* €/xj3aAov(ja vSaro^ icat rot? ^ctXctri TrpoortjiipowTa TrXrjpov ff>ikj}piaTtiiV to iKwmpia Kal ovrta^

StSow Tw ^co/jtcVotf. Elsewhere PhOostratus, whose fancy, like that of Herrick, seems
to have run much on love and roses, plays on the same thoughts {Epist. 60 and 63).
Another passage in his letters {Episf. 55^ papalvtrai koL yvi^r} ptrh p6Bu>Vt av ^paBvv^
Mr) juucAAe, ut KoXiq' <rvpwal^mpi€\\ <Tr€i^av*a<r^p.tBa r€n% poSot^, ^vhpapmpL€v) might have
served as a text to Herrick's

Gather ye roae-hiids while yc may.

But without doubt the English poet drew his inspiration from living roses in
English gardens and English hedges^ not from dead Greek roses in the dusty pages
of Philostratus,

FOLK-LORE IN THE OLD TESTAMENT

151

§ 6. Not to seethe a kid in its mothers milk.

A modem reader is naturally startled when among the solemn
commandments professedly given by God to ancient Israel he finds
the precept : * Thon shalt not seethe a kid in its mother's milk/ *
And his surprise is not lessened but greatly increased by an atten-
tive study of one of the three passages in wliich the command is
recorded ; for the context of the passage seems to show, as some
eminent critics from Goethe downwards have pointed out^ that the
injunction not to seethe a kid in its mother's milk was actually one
of the original Ten Commandments.^ The passage in question occiu-s
in the thirty-fourth chapter of Exodus. In this chapter we read an
account of what purports to be the second revelation to Moses of
the Ten Commandments, after that in his anger at the idolatry
of the Israelites he had broken the tables of stone on which the
first vei-sion of the commandments was written. What is professedly
given us in the chapter is therefore a second edition of the Ten
Commandments. That this is so appears to be put beyond the
reach of doubt by the verses which introduce and which follow
the list of commandments. Thus the chapter begins : * And the
Lord said unto Moses, Hew thee two tables of stone like unto
the first: and I will write upon the tables the words that were
on the first tables, which thou brakest/ ^ Tlien follows an account
of God's interview with Moses on Mount Sinai and of the second
revelation of the commandments. And at the close of the passage
we read : * And the Lord said unto Moses, Write thou these words :
for after the tenor of these words I have made a covenant with thee
and with Israel And he was there with the Lord forty days and
forty nights ; he did neither eat bread, nor drink water. And he
wrote upon the tables the words of the covenant, the ten command-
ments.** Tlius unquestionably the writer of the chapter regarded
the commandments given in it as the Ten Commandmenta

But here a diflSculty arises ; for the commandments recorded in

^ Exodus xjtiii. 19, xxxiv. 26; Deuteronomy xiv. 21.

' Professor Wellhauien reached this conclusion independently before he found
that he had been anticipated by Goethe, See J, Wellhausen, Die Coniposiimi des
HexateucJis und der historischen Bucher des Alien Testaments'^ (Berlin, 1889), pp.
SGsqq., 327-83 ; K. Btidde^ Gesdtichte der aUftebraiscJien LHiemtur (Leipsic, 1906),
pp. 94-6.

* Exodus xxxiv, 1.

* Exodus xxxiv. 27, 28,

152

FOLKLORE IN THE OLD TESTAMENT

this chapter agree only in part with the commandments contained
in the far more familiar version of the Decalogue which we read in
the twentieth chapter of Exodus and again in the fifth chapter of
Deuteronomy. Moreover, in that second version of the Decalogue,
with which we are here concerned, the commandments are not
given with the brevity and precision which characterize the first
version, so that it is less easy to define them exactly. Accordingly
critics have diflfered as to some details in their enumeration of the

* precepts. The following is the enumeration given by Professor
Budde in his recent History of Ancient Hebrew Literature ^ :

L Thou shalt worsliip no other god.

2* Thou shalt make thee no molten gods.

3. All the firstborn are mine.

4r, Six days thou shalt work, but on the seventh day thou
shalt rest.

6. The feast of unleavened bread shalt thou keep in the month
when the com is in ear,-

6. Thou shalt observe the feast of weeks, even of the firstfruita
of wheat harvest, and the feast of ingathering at the year's end.

7. Thou shalt not offer the blood of my sacrifice with leavened
bread.

8. Tlie fat of my feast shall not remain all night until the
moming.3

9. The first of the firstfruits of thy ground thou shalt bring
unto the house of the Lord thy God.

10. Thou shalt not seethe a kid in its mother's milk.

The difference between this version of the Decalogue and the

one with which we are familiar will at once strike the reader.

\. Here morahty is totally absent, Tlie commandments without

exception refer purely to matters of ritual. They are religious in

» the strict sense of the word, for they define with scrupulous, almost

^ J niggling, precision the proper relation of man to God. But of the

^ relations of man to man, not a word. The attitude of God to man

• in these commandments is like that of a feudal lord to his vassals.

* K. Budde, Gesclnchte der aUhehmiscJien Litieraiur, p. 95.

* This commandment does not appear in Exodus xxxlv. but it occurs in the
parallel yersion of the Decalogue in Exodus xxiii. 15,

■ The vereion of the commandment given in Exodus xxiii, 18 is here prefeiTed
to the different version in the pai-allel passage Exodus xxxiv. 26 ; * Neither shall
the aacrific© of the feast of the passover he left unto the morning,*

FOLK-LORE IN THK OLD TESTAMENT

153

He stipulates that they shall render him his dues to the utmost
farthing, but what they do to each other, so long as they do not
interfere with the payment of his feu duties, is no concern of his.
How different from the six concluding conmiandments of the other
version : ' Tliou shalt honour thy father and mother ; thou shalt do
no murder ; thou shalt not commit adultery ; thou shalt not steal ;
thou shalt not bear false witness against thy neighbour ; thou shalt
not covet thy neighbour s house, thou shalt not covet thy neighbour's
wife, nor liis manservant, nor his maidservant, nor his ox, nor his
ass, nor anything that is thy neighbour's/ '

If we ask which of these two discrepant versions of the Deca-
logue is the older, the answer cannot be doubtful It would happily
be contrary to all analogy to suppose that precepts of morality, which
had originally formed part of an ancient code, were aftenvards struck
out of it to make room for precepts concerned with mere points of
ritual. Is it credible that, for example, the command, *Thou shalt
not steal,' was afterwards omitted from the code and its place taken
by the command, ' The fat of my feast shall not remain all night
until the morning'? or that the command, 'Thou shalt do no
murder,' was ousted by the command, * Thou shalt not seethe a kid
in its mother's milk ' ? The whole course of human history refutes
the supposition. All probability is in favour of the view that the
moral version of the Decalogue, if we may call it so from its pre-
dominant element, was later than the ritual version, because the
general trend of civilization has been, still is, and we hope always
will be, towards insisting on the superiority of morality to ritual, \
It was this insistence which lent force to the teacliing, first, of the
Hebrew prophets, and afterwards of Christ himself. We should «
probably not be far wrong in surmising that the change from the
ritual to the moral Decalogue was carried out mider prophetic
influence,^

' Exodus XX. 12-17.

* In assuming the ritual version of the Decalogue (Exodus xxxlv) to be older
than the moral version, I agree with Frofe^ors Wellhausen and Budde (IL cc,).
But in estimating the comparative age of the two versions I purposely leave out of
account the difference of the two documents (the Jehovistic and the Elohistic) in
which they are found, because critics are not agreed as to the relative age of these
two documents. If, however, some of the best critics (including Knenen, Well-
hausen, Stade, and Driver) are right in assigning the priority to the Jehovistic
document, this would be another argument in favour of the earlier date of the
ritual Decalogue (Exodus xxxiv)> since it is Jehovistic ; whereas the moral Decalogue

154

FOLK-LORE IN THE OLD TESTAMENT

But if we may safely assume, as I think we may, that the ritual
version of the Decalogne is the older of the two, we have still to ask
why was the precept not to seethe a kid in its mother s milk deemed
of such vital importance that it was assigned a place in the primitive
code of the Hebrews, while precepts which seem to us infinitely more
important, such as the prohibitions of murder, theft, and adultery,
were excluded from it?' To suppose with some commentators,
ancient and modem, that the precept is one of refined humanity ^ is
in the highest degree improbable. A legislator who, so far as appears
from the rest of the princdtive Decalogue, paid no attention to the
feelings of human beings was not likely to pay much to the maternal
feelings of goats. It is far more probable that the command was
based on some superstitious beUef current among a rude pastoral
people who depended for their subsistence chiefly on their flocks
of goats.

Now among pastoral tribes in Africa at the present day thei*e
appears to be a widely spread and deeply rooted aversion to boil the
milk of their cattle, the avei-sion bemg based on an idea that a cow
whose milk has been boOed will yield no more milk, and that the
animal may even die of the injury thereby done to it. To take
examples. The milk and butter of cows form a laige part of the
diet of the Mohammedan natives of Sierra lioone and the neighbour-
hood J but * they never boil the milk, for fear of causing the cow to
become dry, nor will they sell milk to any one who should practise
it. The Bulloms entertain a similar prejudice respecting oranges,
and will not sell them to those who throw the skins into the fire,
** lest it occasion the unripe fiiiit to fall off/' ' ■ Thus it appears that
the objection to boil milk is based on the principle of sympathetic
magic. Even after the milk has been drawn from the cow it is
supposed to remain in such vital connexion with the animal that

(Exodus xjc) is Elohistic See S. K, Driver, Intmduction to the Literature of the Old
Testatneni^, pp. 29 sq.^ 116 ; id*. Hie Book of Genesis*, p, xvi ; J. Estlin Carpenter
and G. Harford Battersby. The Hemtaich (London, 1900), i. 276, ii, IIL

^ Rol>ert9on Smith thought that the command not to seethe a kid in Its mother's
milk was directed against a form of heathen sacrifice (Eeligion of the Semites \
p. 221 note). But he adduces no example of such a sacrifice^ nor do I remember
to have met with any in my reading.

" See A. Dillmaiin's commentary on Exodus xxiii, 19.

^ Th. Winterbottom, An Acemmt of the Natim Africans in the neighbourhood of
Sierra Leone (London, 1803), pp, 69 sc^.

FOLK-LOEE IN THE OLD TESTAMENT

166

any injury done to the milk will be sympathetically felt by the cow.
Hence to boil the milk in a pot is like boiling it in the cow's udders ;
it is to dry up the fluid at its source.

On the opposite side of Africa we meet with the same supersti-
tion among pastoral peoples. When Speke and Grant were on their
memorable journey from Zanzibar to the source of the Nile, they
passed through the district of Ukuni, which lies to the south of the
Victoria Nyanza. The king of the country lived at the village of
Nunda and * owned three hundred milch cows, yet every day there
was a difficulty about purchasing milk, and we were obUged to boil it
that it might keep, for fear we should have none the following day,
Tliis practice the natives objected to, saying, '* Tlie cows will stop
their milk if you do so." ' ^ Among the Waganda the same rule is
stringently observed, and for the same reason*=^ Similarly the
Bahima, a pastoral people of Enkole in the Uganda Protectorate,
strictly abstain from boiling milk^ They beUeve that to boil it
would cause the cows to fall iU and die.^* They even say that ' if
a Em^opean puts his milk into tea it will kill the cow which gave
the mOk. Also the Bairo, who eat sweet potatoes and ground-nuts,
are not allowed to drink milk, as it would then injure the cattle ; so
in tlie old days before rupees and kauri-shells were introduced butter
was a common currency, but they could not sell the milk itself for
fear that it might be drunk by some one who was forbidden to drink
it/ * In like manner the Masai, who are, or used to be, a purely
pastoral people, regard the boiUng of milk as a crime which they
would neither commit themselves nor allow others to commit.' Tlie
reason for their avei^ion to the practice is not given, but in the light
of the foregoing evidence we may safely assume that they fear to
injure or kill the cows by boiling their milk. The same prohibition

' J. A. Grant, A Walk iicross Africa (Edinburgh and London, 1864), p. 89.
' This I l«arn from my friend the Kev. J. Roscoe, for many years a missionary
in Uganda.

* J* Bo8Co«. * The Bahinia, a cow tiibe of Enkole, in the Uganda Protectorate,'
Journal of tlie Anthropohtjical Imiitute, xxxvii. Ill (1907).

* Major Meldon, * Notes on the Bahima of Ankole/ Journal of the African
Socktt/j No. atxii, January, 1907, p, 142.

* P* Keichard^ DcutscMMafriku (Leipsic. J892), pp, 287 sq. However, milk
mixed with blood and heated is given by them to the wounded* But this practice
is said to have been borrowed from outaide. See 0. Baumann, JJttrch Massailand
Mur Mlquelk (Berlin, 1894), p. 162. Compare M. Merker, Die Mami (Berlin, 1904),

p. a2.

158

FOLK^LORE IN THE OLD TESTAMENT

to boil milk is observed also by the Wagogo, the Wamegi^ and the
Wahumba, three tribes of Grermaii East Africa/'

A similar fear of tampering ^th the principal source of subsis-
tence may well have dictated the old Hebrew commandment : - Tliou
shalt not seethe a kid in its mothers milk/ On this theory an
objection ^vill be felt to seething or boiling a kid in any mUk, because
the she-goat from which the milk had been drawn would be injured
by the process, whether she was the dam of the kid or not. The
reason why the mother s milk is specially mentioned rather than
milk in general may have been either because as a matter of
convenience the mother's milk was more likely to be used than
any other for that purpose, or because the injury to the she-goat in
such a case was deemed to be even more certain than in any other.
For being linked to the contents of the boiling pot by a double bond
of sympathy, since the kid, as well as the milk, had come from
her bowels, the mother goat was twice as likely as any other goat
to lose her milk or to be killed outright by the heat and ebidlition.

But it may be said : If the objection was simply to the boiling
of milk, why is the kid mentioned at all in the commandment ?
The practice^ if not the theoiy, of the Baganda seems to supply the
answer. Among these people it is recognized that flash boiled in
milk is a great dainty, and naughty boys and other unprincipled
persons, who think more of their own pleasure than of the welfare
of the herds, will gratify their sinful lusts by eating meat boiled in
niUk, whenever they can do so on the sly,^ heedless of the sufferings
which their illicit banquet inflicts on the poor cows and goats.
Thus the Hebrew command *Thou shalt not seethe a kid in its
mother's milk' was probably directed against miscreants of this
sort, whose surreptitious joys were condemned by pubhc opinion as
striking a fatal blow at the staple food of the community. We can
therefore undemtand why in the eyes of a primitive pastoral people
the boiling of milk should seem a blacker crime than robbery and

' This I learn from my friend the Rev, J. Roseoe, whose information is based
on personal contact with all three tribes. However, the prohibition to boil mOk is
not universal among pastoral people, Thns among the Wataturu of East Africa, who
used to live mainly on flesh and milk, the practice of hoiling milk was always quite
common. See 0, Baumann, Durdi Masstiiiand sur N'dquelk (Berlin, 1894), p. 171.
And the modern Bedouin of Arabia seem to \m\ milk without scruple. See
J* L. Burckhardt, Notes on (Jw Bedouin and Wahabtfs, u 63 ; C. M, Doughty, Travels
in Ambia Deseiia, il 67.

* So I am told by my friend the Rev. J. Koscoe.

FOLK-LORE IN THE OLD TESTAMENT

157

murder. For whereas robbery and murder harm only individuals, '
the boiUng of milk, like the poisoning of wells, seems to threaten the
existence of the whole tribe by cutting off its principal source of
nourishment. That may be why in the first edition of the Hebrew
Decalogue we miss the commandments * Thou shalt not steal ' and
*Thou shalt do no murder', and find instead the commandment
' Tliou shalt not boil milk/

The conception of a sympathetic bond between an animal and •
the milk that has been drawn from it appears to explain certain
other rules observed by pastoral peoples, for some of which no
sufficient explanation has yet been suggested. Thus milk is the
staple food of the Damaras or Hereros of south-western Africa,
but they never cleanse the milk-vessels out of which they drink
or eat, because they believe that were they to wash out the vessels
the cows would cease to give milk.^ Apparently their notion is
that to wash out the lees of the milk from the pot would be to
wash out the dregs of tlie milk from the cow's udders.

Again, it is a rule with the Caffre tribes of South Africa and with •
the Bahima of Enkole that menstruous women may not drink milk ;
and the reasons assigned for the rule prove that the idea on which
it rests is the supposed sympathy between the milk and the animal.
Thus among the Balinna a woman at her montlily periods must
eat vegetables and drink beer ; for it is thought that if she drank
milk she would thereby injure the cows. But an exception is
made m favour of a girl at her first menstruation ; her father sets
apart for her use an old cow, the milk of which is her only food.*
The exception is significant. An old cow will soon lose her milk
in any case, so it does not signify much if she loses it a little sooner
through the pollution of her milk by the menstruous girl. The
Cafires of South Africa believe that the cows would die if a men-
struous woman tasted their milk.^ Even the maidens who attend

^ C. J. ADderssoB, Lake Nt^mi ', p. 230 ; J. Hahn, * Die Ovaherero,' Zeitschi/t
dcr GmllBchaj} fur Erdkunde £u Berlin, iv (1869), p. 250,

^ J. Roscoe, * The Bahima, a cow tribe of Enkole in the Uganda Protectorate/
Journal of the Anthropolofiical In^itute, xxxvii. 107 (1907).

* J. Mocdonald. * Manners, customs, superstitions and religions of South African
tribes/ Journal of (he Atfihrf^poloffical Institute^ xx (1891), p, 138; *ff., Light in
Africa^, p. 221. Compare L. Alljerti, De A'^/ers (Amsteidani, 1810), pp* 102 sq.
For a similar reason, probably, among the Bacas of South Africa a woman at
nienfitniation is not allowed to see or touch cow's dung (J. Macdonald, in Journal
qfthc Anthropological Institute^ xx (1891), p. 119).

158

FOLK-LORE IN THE OLD TESTAMENT

a girl at her first menstruation are forbidden to drink milk, lest the
cattle should die ; the period of seclusion and taboo lasta a fort-
night.^ If a Caffre woman infringes tliis custom at one of her
periods her husband may be fined from one to three head of
cattle, which are paid to the chief. Formerly this monthly period
of abstinence from milk lasted for seven or eight days,'^' Further,
among the Caflfres menstruous women are forbidden to cross those
parts of the kraal which are frequented by the cattle ; for if a drop
of their blood were to faU on such a path * any oxen passing over
it would run great risk of dying from disease'. Hence %vomen
have to make circuitous paths from one hut to another, going
round the back of the huts in order to avoid the forbidden ground.
The tracks which they use may be seen at every kraaL But there
is no such restriction on the walks of women who are past child*
bearing^ because they have ceased to be a source of danger. *

The disabilities thus imposed on women at menstruation are
perhaps dictated by a fear lest the cows whose milk they drank
should yield milk mingled with blood. Such a fear, Mr. Eoscoe
tells me, is much felt by the pastoral tribes of Central Africa,
Again, the same idea perhaps explains the Zulu custom which
forbids a wounded man to drink milk until he has performed
a certain ceremony. Thus when an Englishman serving with the
Zulus was wounded in action and bled profusely, a young heifer
was killed by order of the medicine-man^ and its small entrails,
mixed with the gall and some roots, were parboiled and given to
the sufferer to drink. At fii-st he refused the nauseous dose, but
the medicine*man flew into a passion and said * that unless I drank
of the mixture I could not be peraiitted to take milk, fearing the
cows might die, and if I approached the king I should make him
illV* This fear of injuring the cows through the infection of blood

* Dudley Kidd, The Esseniial Kitfir, p. 209.

' Com^ndium of Kajir Laws and Customs (Cape Town, 1866), p. 122, comjwii^e
p. 9t.

^ Dudley Kidd, Tfw Essential Kafir, pp. 2S8sq. ; Compenflium of Kafir Laws
a/nd Customs, p. 93. The huts of a Caffre kraal are usually arranged in a circle
with the cattle fold in the centre (Dudley Kidd, op* cit,, pp, 12 sq*)* Hence the
women *s paths may be supposed to He outside the circle of the huts, between them
and the palisade which sometimes encloses the kraaL

. * Nathaniel Isaacs, Travels and Adventures in Eastern AJrka (London, 18B6), i.
20a-6.

POLK-IiORE IN THE OLD TESTAMENT

159

may perhaps explain a Bechuana custom of removing all wounded
persons to a distance from their towns and villages,^

The same dread probably lies at the root of the stringent rule
which among many Afiican tribes, especially of the Bantu family,
forbids women to milk the cows and to enter the cattle-fold,^' But
if for some reason a married Caffre woman is obliged to enter
a cattle-fold she must bring her husband or her nearest male
relative to the gate of the fold ; there he lays his spear on the
ground with the point inside of the entrance, and the woman walks
in on the handle of the weapon. * This is considered as a passport
of entrance, and saves her fi'om punishment : but, even in this
case, strict inquiiy is made as to the necessity for such an entrance,
nor are the men very willing to grant, too frequently, such an
indulgence to them/^ Amongst the pastoral Todas in southern
India the business of milking the cattle is performed by men only,
who are invested according to their rank with various degrees of
sanctity and have to observe strict rules of ceremonial purity.
Toda women take no part in the ritual of the sacred dairy nor in
the operations of milking and churning which are there carried
on. They may go to the dairy to fetch buttermilk, but they must
approach it by an appointed path and stand at an appointed place
t43 receive tlie milk. Only under very special conditions is a woman
or a girl permitted to enter a dairy. Indeed, during the performance

' R. Moflfat, Mmionafy Labours and Scenes in Southern Africa (London, 1842),
p» 465. Dr. Moffat could not aaeertain the reason of the custom^

» Dudley Kidd, The Essential Kafir, p. 238 ; J, Campbell, Travels in South
Africa^ Second Jmmey (London, 1822), n* 21S ; E. Casalis, The Basutos^ p. 125 ;
Compendium of Kafir Laws and Customs, p, 93 ; F. Fleming, Southern Africa (London,
1866), pp* 214 aq. ; id,, Kajfforia (London^ 1853), p, 98 ; Krantz, Natur- imd Kidfur-
hhen der Zulus^ pp. 81 sq. ; J, Macdonald, Liffkt in Africa*, p. 221 j F. Lichtenstein,
Eeisen im Siidlichen Affiku^ L 441; H. Schinz, Lkutsch'Stid-West-Afrikaf p. 296;
L. Grout, Zululand, p. Ill ; J. Mackenzie, Ten Years North of the Orange Eiver,
p, 499 ; G- Fritsche, Die Eingehorenen SUd-Afrikas, pp. 85, 183 ; Emin Pasha in
antral Africa (London, 1888), pp, 238, 343; Sir H. H. Johnston. British
Gmtral Africa, p. 431 ; C. T. Wilson and R W. Felkin, Uganda atid the
Egtfptian Soudan (London, 1882), I 164 ; R. W. Felkin, * Notes on the Madi or
Mom tribe of Central Africa/ Proceedings of the Rogal Society of Edinbfirgh, xii.
(1882-^), pp. 306 Bq.; H, Cole, * Notes on the Wagogo of German East Africa/
Journal of the Anthropological Institute, xxxii (1902), p. 337 ; W, Munzinger,
Sitten und Becht der Bogos, pp. 77 sq.; id,, Ostafrikanische Studien, p. 325. How-
ever, it deserves to be noticed that among the Bechuanas, while cows are always
milked by men, goats are always milked by women {J. Campbell^ loc. cit*),

' F. Fleming, Southern Africa, pp. 214 sq,

160 FOLK-LORE IN THE OLD TESTAMENT

of certain ceremonies at the dairy women are obliged to leave the
village altogether.*

However, this sedulous seclusion of women from cattle is not
practised by all pastoral tribes. For instance, the cows are milked
by women among the Hottentots, Korannas, and Hereros of South
Africa , among the Masai of East Africa \ and among the Dinkas
of the Upper Nile/ So far indeed are the Namaquas, a Hottentot
tribCj from sharing the superstition as to the disastrous influence
of menstruous women on milk and cattle that among them, when
a girl attains to puberty, she is led round the village to touch the
milk-vessels in the houses and the rams in the folds for good luck/
With this custom we may compare a practice of the Hereros. Among
them the fresh milk of the cows is brought by the women to the
chief or the owner of the kraal, at the sacred hearth or sacrificial
altar, and he tastes and thereby hallows the milk before it may be
converted into curds. But if there happens to be a lying-in woman
in the kraal, all the fresh milk is taken to her, and she consecrates
it in like manner instead of the chief/ Among the Bedouin of
Arabia the mileb camels are milked by men and lads only, but
the sheep and goats are milked by women. ^ Among the Calmucks
of Siberia it is the business of the women to milk the cattle/
and among the Lapps the reindeer are milked by men and women
indifferently.*

The pollution of death is also with some people a bar to the
drinking of milk. Thus, when a death has taken place in a Zulu

' W. H. R. Rivera, The Todas (London, 1906), pp. SOsqq., SSsqq., 281 gqq.,
especially 245 sq.

• P. Kolben, TM Present State of the Cape of Good Hope (London, 178S), i. 171,
172 ; J. Mackenzie, Ten Tears North of the Orange Blmr, p. 499; J, Irle, Die Hereto
(Gtltersloh, 1906), p. 121. Among the Hottentots the milk of cows is drunk by
both sexes, but the milk of ewes only by women (F. Kolben, op, cit*, i. 175).

» A- C. HoUls, Tfm Masai (Oxford, 1905), p. 290.

• £min Pasftu in Central Africa, p. 348.

* Sir J. E. Alexander, Expeditkm ofJDi&covery into the Interior of Africa (London^
1838), i, 169.

* Rev. E. Dannert, * Customs of the Ovaherero at the birth of a chOd *
{South African) Ihlk-lore Journal iL 63 sq. ; J. Irle, Die Hercro (Gtltersloh, 1006)^
ppw 79, 94.

' €. M. Doughty, Travels in Arabia Bc^ertafh 261 eq. ; J. L. Burckhardt, Notes
on the Bedouins and Wahab^s^ i. 239.

* R S. Pallas^ Eeisc durch verschiedene Prommen des RussiscJien Beichs^ i. 314,

• J. SchefFer, La]^onia (Frankfort, 1673), p. 331.

FOLK-LOKE IN THE OLD TESTAMENT

161

village, no milk is drunk nor are the cattle allowed to be milked
on that day.^ And with regard to the Caffres of South Africa in
general we are told that no person ceremonially unclean may drink
milk, and that among such persons are a widow and a widower,
the widow being unclean for a month and the widower for half
a month after the death of husband or wife respectively/- Similarly
among the Todas a widower and a widow are forbidden to drink
milk for a period which may extend to many months.^ The reasons .
for these prohibitions are not given, but in the Ught of the foregoing
evidence we may conjecture that the motive is a fear lest the cows
might die if their milk were drunk by a man or woman who was
thus deeply tainted with the pollution of death. Yet in apparent •
contradiction with this fear is the treatment of a widow among
the Bechuanas, ' When a woman's husband is dead, she may not
enter a town, unless she has been under the hands of a sorcerer.
She must remain at some distance from the town ; then a little
milk from every cow is taken to her, which mixture of milk she
muBt boil with her food. Dung from the cattle pens is also taken
to her, and with this, mixed with some nmlenw^ she must rub herself.
If this ceremony be not gone through, it is thought that all the
cattle in the town will surely die/* How these ceremonies are
supposed to prevent the cattle from dying, I do not see ; but at least
it appears that the milk and the dung of the cows are both beUeved to
remain in sympathetic connexion with the animals, since the use of
them by the widow is supposed to save the cattle aUve. Under certain *
circumstances maternity as well as death is thought to endanger
the herds. In the Nandi tribe of Eastern Uganda, when a woman
has given birth to twins, she has to Uve apart for some months,
and may not go near the cattle fold ; for if she did, they think that
the cattle would die. But one cow is put aside for her, and she
drinks its milk/ Another curious example of sympathetic magic

* A* F. Gardiner, Narrative of a Journey to tfte Zoolu Country (Londoo, 18»36),
p. 81.

* L. Alberti, De Kaffers aan de Zuidkmi van Afirika (Amsterdam, 1810),
pp. 102 eq.

* W. H, E. Rivers, The Todas, p. 24 L

* Mias J, P. Meeuwesen, ' Customs and SiiperstitioDB among the Betshuana/
{South African) Folk-lore Joumul, i (1879), p, 34. The word mdem^ means both
poison and medicine.

^ C. W. Hobley, Eastern Uganda (London, 1902), pp. 39 sq,

TTbOR K

162

FOLK-LORE IN THE OLD TESTAMENT

applied to the milk of cattle may be mentioned, though it does not
fall in with the other instances which I have cited. The Kabyles
of North Africa believe that whoever gets possession of the herdS'
man's staff can conjure the milk of that herd into the udders of his
own cows. Hence when he retires to his house in the heat of the
day, a herdsman takes care not to let go his staff for a moment.
To sell the staff or allow another to get hold of it doling the siesta
is an offence which is punished with a fine.^

Among the Wakamba and Wakikuyu of Central Africa inter-
course between the human sexes is forbidden so long as the cattle
are at pasture, that is, from the time when the herds are driven
out in the morning till the time when they are driven home in
the evening.^ The reason for this prohibition is not mentioned j
but we may conjecture that the intercourse of the sexes is supposed
to be in some way injurious to the cattle while they are at grass.
For a similar reason, perhaps, the most sacred dairymen of the
Todas must avoid women altogether.^ An idea of the same sort
may underlie the Caflfre custom which restricts the use of fresh
milk to young people and very old people ; all other persons, that
is, all adults in the vigour of life, may only use curdled milk,*
Among the Bechuanas * there are two months in the year, at the
cow-calving time, which is generally about the month of October,
when none but the uncircumcised are permitted to use the milk of
cows that have calved'.^ As the uncircumcised would usuaEy be
under puberty, it seems Ukely that this Bechuana rule is in some
way based on the idea that under certain circumstances the inter-
course of the human sexes may injuriously affect the cattle. Perhaps
the practice of eating milk in the form of sour curds, which prevails

* X Lioret KabpUe du Jutjura (Paris, n. d.), p. 512.
' J. M. Hildebmndt, ' Ethnographiache Notlzen iXhsr die W&kamba unci ihre

Nachbarn/ Zeitschrift fur Ethnohgie, x {1878), p. 40L
' W. H. R. Rivers, The Todas, p. 2B6.

* J. Shooter, The Kafirs of Natal and tite Zulu countr^j p. 28, Similarly
Mr. Dudley Kidd writes : ' Sweet milk is but food for babies^ and only a few tribes
would drink it. But clotted sour milk is food for men ' {TJte Essential Kafir, p. 59).
* In the south of Africa it is only the children who drink milk in a sweet state'
(E. Casalis, The BasidoSf p. 145). Again, in the Kikuyu tribe of British East
Africa the milk both of cows and goats is much used^ but only children drink it
fresh (H. R Tate, * Further Not€>s on the Kikuyu tribe/ Journal of the Anthrvpological
Institute, xjtxiv (1904), p. 259),

* J. Campbell^ Travels in South Africa, Second Jmmey^ ii» 202.

FOLK-LORE IN THE OLD TESTAMENT

163

among the pastoral tribes of Africa,^ may spring not altogether
from a preference for curds, nor yet fiom the difficulty of keeping
milk fresh in a hot climate^ but partly at least from a superstition
that the sympathGtic bond between the cow and its milk is weakened
or severed when the milk has been turned into curds or buttermilk,
and tliat accordingly you run less risk of sympathetically hurting
the cow when you eat curds than when you drink fresh milk.
Some such idea at all events would explain why in the cases just
cited the drinking of fresh milk is confined to the young and the
old, that is. to the classes who are physically unable to endanger
the supply of the precious fluid in the manner indicated. The
Baliima seem to suppose that the sympathetic bond between the
milk and the cow is severed when the milk is converted into butter;
for, whereas they will not sell the milk lest it should fall into
the hands of persons who would injure the cows by drinkuig it,
they never had any objection to parting with butter,^ From all
this it appears that any process which converts milk into another
substance, such as curds, butter, or cheese, may be regarded, though
it need not necessarily be regarded, as snapping tlie link which
binds the milk to the cow, and therefore as enabling the milk in
its new form to be used by the profane without injury to the cattle. *

' F, Fleming, Kaffraria, pp* 108 sq, ; id., Southern AJ'rica, pp. 218 sq.; K
Alberti, De Kaffers aan de Zttidhtst van Afrikay p. 86 ; H. R. Tate, * Further Notes
on the Kikuyii tribe of British Exist Africa, Vournai of the Anihropologiml Institute,
xxxiv (1904). p. 259 ; Sir H. H* Johnston, British Central Africa, p, 431 ; Dudley
Kidd, Hie Essential Kqfir, p. 59 ; E. Casalis, T/w; Basuios, p. 146 ; E. Dannert,
* Customs of the Ovaberero.' (South African) Folk-hre Journal^ ii 63 ; R Speckmami,
Die Hermann shurger Mission in Afrika (Hermannsburg, 1876), pp. 107 sq. The
process of making the curds is thus described by Mr. Kidd (loc, cit): 'When the
milking is over the milk is taken into the hut, and is immediately placed in the
milk sac or calabash. This is never cleaned outj but contains a strong ferment
which makes the milk clot immediately. * . . The calabash has a small plug at the
bottom by which the natives let off the whey, the curds being the only pait they
care for/ On the other hand, the Masai drink milk both fresh and sour (M.
Merker, Die Masai p. 32), and the Bahima drink it only fresh (J. Eoscoe, ' The
Babima/ Jmirmd of the Anthropdogical Institute, xxxvii. 100)* The Bedouin of
Arabia * drink no whole-milk save that of their camels ; of their small cattle they
drink but the butter- milk ' (C. M. Doughty, Travds in Arabia Beserta, h 825).

' See above, p. 155.

* When the Wanyamwesi are about to convert milk into butter, they mix it
with the urine of cows or of human b«ings. The reason they gave to Stuhlmann
for this practice was that it made the butter more saleable ; but he believed,

M 2

164

FOLK-LORE ^N THE OLD TESTAMENT

Among tribes who hold such views the operations of the dairy aim,
so to say, at disenchanting the milk for the l>enefit of the cow, at
breaking the tie which binds the two together lest it should di-ag
the animal down to death,

^ Lastly, the supposed sympathetic influence of milk on the

cow is the reason why the Masai take the utmost pains not to
bring milk and flesh into contact with each other ; for they imagine

I that contact between the two would set up a disease in the ndders
of the cow from which the milk was drawn. Hence it is a rule

/ with them never to keep flesh and milk in the same vessel ; different
sets of vessels are set apart for the one and the other. For the
same reason they seldom can be induced to sell their milk, lest
the purchaser should make their cows ill by bringing it into
contact with flesh. Hence, too, Masai waixiors will not eat flesh
and milk on the same day. Their practice is to eat nothing but
milk for some days and then nothing but flesh and blood for several
days more. But before they pass from the one diet to the other
they take a strong purgative in order to make sure that no vestige
of the previous food remains in their stomachs ; so scrupulous are
they not to bring milk into contact with flesh or blood. They
think that if they failed to observe this precaution the cows would
give less milk. Moreover, even when they do eat flesh and drink
blood, they may not do so in the kraal ; they must retire to a
lonely place in the forest, there to kill a bollock and gorge them-
selves on its flesh and blood. The reason for this particular rule
may perhaps be, either wholly or in part, a delicate wish to spare
their cattle the pain of witnessing the slaughter and consumption
of their fellows. Further, the use of game, and especially of corn
of all sorts, is strictly forbidden to the Masai warrior ; if he ate
corn he would get no wife. Besides flesh, blood, and milk the
warriors may eat only honey and sugar-cane.^ The reason for

probably with justice, that the real motive was a fear that the cows would lose
their milk if this procedure were not followed. The Wanyamwesi do not eat the
milk thus polluted ; they only use it to smear oo their persons. See F. StuMmann^
MU Eniin Pasdia hts Herz von Afriku^ pp, 78 sq»

' J, Thomson, Through Masai Land (London, 1885), pp. 429->ai j P. Keichard,
Deutsch'Ostafriht {Lelpme, 1802), pp. 287 eq* ; O, Baumann, Durch 3l($$sailund sur
Nilqttelk {Beilin, 1894), pp. 161 sq. ; M. Merker, Die Masai (Berlin, 1904), p. 03.
Only the last of these writers mentions the supposed sympathetic connexion of the
milk with the cows as the reason for the taboo. Among the Watatnru of East
Africa any man who ate of a certain species of antelope (called in Swahili ponu) was
formerly ibrbidden to drink milk on the same day (0. Baumann, op, ciL^ p, 171).

FOLK-LORE IN THE OLD TESTAMENT

165

the embargo thus laid on game and com is not mentioned ; but on
the analogy of the former taboo we may surmise that the motive
is a fear of injuring the cows in some way by bringing their milk
into contact with these viands.

Similar, though somewhat less stringent, rules as to the sepa*
ration of flesh and milk are observed by the Israelites to this day,
A Jew who has eaten flesh or broth ought not to taste cheese oj
anything made of milk for an hour afterwards ; straitlaced people
extend the period of abstinence to six hours. Moreover, flesh and
milk are carefully kept apart. Thei^ are separate sets of vessels
for them, each bearing a special mark, and a vessel used to hold
milk must not be used to hold flesh, THvo sets of knives aie also
kept, one for cutting flesh, the other for cutting cheese and fish.
Moreover, flesh and milk are not cooked in the oven together nor
placed on the table at the same time; even the table-cloths on
which they are set ought to be different. If a family is too poor
to have two table-cloths, they should at least wash their solitaiy
table-cloth before putting milk on it after meat.' These rules, on
which Rabbinical subtlety has embroidered a vax-iety of fine dis-
tinctions, are professedly derived from the commandment not to
seethe a kid in its mother's milk. Taken all together they have
probably come down from a time when the forefathers of the
Hebrews were goatherds subsisting mainly on the milk of their flocks,
and as afraid of diminishing the supply of it as aie tliose pastoral
tribes of Africa whose supei'stitions on that subject the Jews share to
this day.

The whole of the rules as to the drinking of milk which have
come before us appear to aim at protecting the cows from the hai-ni
which an improper use of their milk is supposed to entail on the
animals ; there seems to be no thought that the wrong act will
directly haim the drinker. It is the cows, and not the people, who are
the immediate objects of the lawgiver's solicitude, if we may speak of
a lawgiver among tribes where custom takes the place of legislation.
Hence we may surmise that the elaborate ritual with which, for
example, the Todas of southern India have fenced the operations of
the dairy was originally designed in like manner for the protection
of the cows ratlier than of the people ; the intention, if I am right,

' J. Buxtorf, Synagoga Judaim (B&le, 1661), pp. 594-6 j J. C. G. Bodenschatz,
KircMiehe Ver/assun^f der heutigen Judm (ErlangeD, 1748), Theil iv. cap. ii.
pp. 25 sq.

166

folk-lore in the old testament

was not so much to remove a taboo from a sacred fluid for the
benefit of the people ^ as to impose a series of restrictions on its
use for the benefit of the cattle. The aim of the ritual was, in short,
to ensure that the cattle should not be injured sympathetically
through the drinking of their milk by improper persons. That the
Todas beUeve such injury to be possible appears from a remark
made by a Toda to a missionary. Ha\dng ascertained the names of
the Toda deities, the missionary was cited to appear before a head-
man to explain how he had come by the information. * I told him,
that as he had no authority to judge me, I should not answer his
question, to which he implied : that I had been drinking the milk of
their buffaloes, on which account many of them would die/ ^ This
answer seems to imply that the milk of the buffaloes remained
in such a sympathetic connexion with the animals that the mere
drinking of it by a stranger might cause their death. The implica-
tion agrees with the express beUefs of pastoral tribes in Africa.

Surveyed as a whole, the evidence suggests that many rites
which have hitherto been interpreted as a worship of cattle may
have been in origin, if not always, nothing but a series of precautions,
based on the theory of sympathetic magic, for the protection of the
herds from the dangers that would threaten them tlirough an indis-
criminate use of their milk by everybody, whether clean or unclean,

' whether friend or foe. The savage who believes that he himself can
be magically injured through the secretions of his body naturally
applies the same theory to his cattle and takes the same sort of steps

^ to safeguard them as to safeguard himself. If this view is right, the
superstitious restrictions imposed on the use of milk which have
come before us are analogous to the superstitious precautions which
the savage takes with regard to the disposal of his shorn liair,
clipped nails, and other severed parts of his person. In their essence

' they are not religious but magical. Yet in time such taboos might
easily receive a religious interpretation and merge into a true worship
of cattle. For whOe the logical distinction between magic and religion
is sharp as a knife-edge, there is no such sharp line of cleavage
between them historically. With the vagueness characteristic of
popular thought the two are constantly fusing with each other, like
two streams, one of blue and one of yellow water, wliich meet and
blend into a river that is neither wholly yellow nor wholly blue.

* As Dr. Rivers seemB to think [The Todas^ pp. 231 sq.).

' F. Metz. The Tribes mhabiting the NeUgherrtf Hilh (Mangalore, 1864), p. 43.

FOLK-LORE IN THE OLD TESTAMENT

167

But the historical confusion of magic and religion no more dispenses
the philosophic student fram the need of resolving the compound
into its constituent parts than the occurrence of most chemical
elements in combination dispenses the chemist from the need of
distinguisliing them. The mind has its chemistry as well as the
body. Its elements may be more subtle and mercurial, yet even
here a fine instrument will seize and mark distinctions which might
elude a coarser handling.

§ 7. TU Keepers of the ThresMd.

In the temple at Jerusalem there were three oflScials, apparently ^^^x
priests, who bore the title of Keepers of the Threshold.^ What
precisely was their function? They may have been mere door- .
keepers^ but their title suggests that they were something more ; for
many curious supei^titions have gathered round the threshold in
ancient and in modern times. The prophet Zephaniah repi-esents
Jehovah himself saying : * And in that day I wUl punish all those
that leap on the threshold, wliich fill their masters house with
violence and deceit/ -^ From this denunciation it would appear that *
to jump on a threshold was viewed as a sin which equally with
violence and deceit drew down the divine anger on the jumper.
At Ashdod the Philistine god Dagon clearly took a similar view *
of the sinfulness of such jumps, for we read that his priests and
worshippers were careful not to tread on the threshold when they

^ Jeremiah xxxy. 4, Hi, 24; 2 Kings xfl 9^ xxii. 4, xxiii. 4, xxv. 18. In all these
paasaged the Etiglish Veraioo, both Authorized and Revised, wrongly subfltitutes
* door ' for * threshold '. The number of these officials is mentioned in Jeremiah
Hi. 24 and 2 Kings xxv, 18. That they were priests seems to follow from
2 Kings xii. 9.

' Zephaniah i. 9. The Kevised Version wrongly renders ^ over the threshold '.
The phrase is rightly translated in the Authorized Version. The English Re-
visers and Kautsch in his German translation of the Bible have done violence
to the proper sense of the preposition 7^ (*upon*)^ apparently for the purpose
of harmonizing the passage with 1 Samuel v. 5* Professor 8* R Driver also
thinks that the prophet is denouncing a heathen practice of Jumping over the
threshold (note on Zephaniah i, % in The Century Bible), and Professor R. H.
Kennett writes to me that he inclines to take the same view* It might be a nice
question of casuistry to decide whether a jumper who clears a threshold has
committed a more or a less deudly sin than one who lights on the top of it In
either case many people will find it hard to understand the indignation of the
Deity on the subject.

168

FOLK-LORE IN THE OLD TESTAMENT

entered his temple.^ The same scruple has persisted in the same
regions to this day* Captain Conder tells us of a Syrian belief * that

. it is unlucky to tread on a threshold. In all mosques a %vooden bar
at the door obliges those who enter to stride across the sill, and the
same custom is observed in the mstic shrines.' ^ Similarly in Fiji
* to sit on the threshold of a temple is tahu to any but a chief of the
highest rank. All are careful not to tread on the threshold of a place
set apart for the gods ; persons of rank stride over ; othei-s pass over
on their hands and knees. The same form is observed in crossing
the threshold of a chiefs house. Indeed, there is very little differ-
ence between a chief of high rank and one of the second order of
deities. The fonner I'egards himself very much as a god, and is ofteix
spoken of as such by his people, and, on some occasions, claims for

* liimself publicly the right of divinity,' ^ Again, when Marco Polo
visited the palace at Peking in the daj^s of the famous Kublai Khan,
he found that * at every door of the hall (or^ indeed, wherever the
Emperor may be) there stand a couple of big men like giants, one

. on each side, armed with staves. Their business is to see that no
one steps upon the threshold in entering^ and if this does happen
they strip tlie offender of his clothes, and he miLst pay a forfeit to
have them back again ; or in lieu of taking liis clothes they give him
a certain number of blows. If they are foreignei-s ignorant of the
order, then there are Barons appointed to introduce them, and

% explain it to them. They think, in fact, that it brings bad luck if
any one touches the threshold. Howboit, they are not expected to
stick at this in going forth again, for at that time some are like to be
the worse for liquor and incapable of looking to their steps/ ^ From
the account of Friar Odoric, who travelled in the East in the early
part of the thirteenth century, it woidd appear that sometimes these
Keepers of the Threshold at Peking gave offenders no choice, but laid
on heartily with their staves whenever a man was unlucky enough to

' 1 Samuel v. 5*

■ C. R* Conder, Ildh and Moab (London, 1883), pp. 293 sq. With regard to
the rustic flhrines, the supposed tombs of saints (above, pp. 116 sqq.), the same writer
observes {TctU Work in Pakstine. IL 221) : * The greatest respect is shown to the
chapel, where the invisible presence of the saint is supposed always to abide. The
peasant removes his shoes before entering, and takes care not to tread on the
threshold/

» Th. Williams, Fiji and the Jflfjioits » {London, 1860), i. 238.

' The Book ofSer Marco Pdo, translated by Colonel H. Yule^ (London, 1875),
L386.

FOLK-LORE IN THE OLD TESTAMENT

169

touch the thresholds When the monk de Rubniquis, who went as
ambassador to China for Louis IX, was at the court of Mangu-Khan,
one of his companions happened to stumble at the threshold in going
out. The wardei's at once seized the dehnquent and caused him to
be carried before * the Bulgai, who is the chancellor, or secretary of
the court, who judgeth those that are arraigned of life and death '.
Howeverj on learning that the offence liad been committed in
ignorance, the chancellor pardoned the culprit, but would never
afterwards let him enter any of the houses of Mangu-Khan.^ The
monk was lucky to get off with a whole skin. Even sore bones were ►
by no means the worst that could happen to a man under these
circiunstances in that part of the world* Piano Carpini, who •
travelled in Tartary about the middle of the thirteenth century,
a few years before the embassy of Rubruquis, tells us that any one
who touched the threshold of the hut or tent of a Taiiiar prince used
to be dragged out through a hole made for the purpose under the
hut or tent and then put to death without mercy, ^ When the
Italian traveller Pietro deUa Valle visited the palace of the Pei-sian
kings at Ispahan, he observed that ' the utmost reverence is shown
to the gate of entrance, so much so, that no one presumes to tread
on a certain step of wood in it somewhat elevated, but, on the
contrary, people kiss it occasionally as a precious and holy tiling/
Any criminal who succeeded in passing this threshold and en-i
tering the palace was in sanctuary and might not be molested.
When Pietro della Valle was in Ispahan, there was a man of rank
living in the palace whom the king wished to put to death. But the
offender had been quick enough to enter the palace and there he was
safe from every violence, though had he made a step outside of the
gate he would instantly have been cut down. * None is refused .
admittance to the palace, but on passing the threshold, wliich he
kisses, as I have before remai^ked, he has claim of protection. This
threshold, in short, is in such veneration that its name of Astaiie is
the denomination for the court and the Royal palace itself.' * Again,

' Colonel H. Yule, Catliatf and tfie Tl% ihitlwr (Rtikluyi Society, 1866). j. 132.
The friar's travels began between 1216 and 1218 and ended in 12S0.

- * Travels of William de Rubruciuis/ Pinkerton*8 Voyages and Travda,
TO. 65-7.

* Jean du Plan de Carpin, delation des Momjoks on Ibriures, ed, D'Avezac
(Parifl, 1888), cap. iii § 2.

* Pietro della Valle, * Travels in Persia/ Pinkerton's Voyages and Travels,
ix. 26, m.

170

FOLK-LORE IN THE OLD TESTAMENT

the Caliphs of Bagdad ^ obliged all those who entered their palace to
prostrate themselves on the threshold of the gate, where they had
inlaid a piece of the black stone of the temple at Meccah, in order
to render it more venerable to the peoples who liad been accustomed
to press their foreheads against it. The threshold was of some
height J and it would have been a crime to set foot upon it* ^

fThus Jehovah's dislike of people who trod on thresholds was
not a Jewish eccentricity, for it has been shared by Fijian chiefe,
Chinese emperors^ Tartar khans, Shahs of Persia, and Caliphs of
Bagdad, as well as by many persons in a humbler walk of life. The
Korwa of north-western India, for example, will not touch the
threshold of a house on entering or leaving it.^ There is a Mongol
proverb : * Step not on the threshold ; it is sin I ' ^ It was a rule of
ancient India that a bride should cross the threshold of her husband s
• house with the right foot firstj but should not stand on it.* In the
Altmark an old German custom required that on her arrival at her
new home a bride should be carried by her husband from the
wagon to the hearth of the house without being allowed to touch
\ the earth with her feet,'* The ancient Roman practice of lifting a
bride over tlie threshold of the bridegroom s house was no doubt
merely an instance of the same widespread superstition ; it had
nothing to do with a practice of marriage by capture, though it has
often been so interpreted from Plutarch's time downward,*^ The
learned Varro more justly derived the custom from the sanctity of
the threshold, "^ and the same view has been rightly taken by some

* D*Herbelot, BiUiQtMque OrientaJe, i (The Hague, 1727), p. 306, a. v, *Bab/
citing as his autiioritj Kliondemir, in the Life of Mostaasem.

' W; Crooke, The Tribes and Castes of the North-Western Proi^inces and Oudh,
iii. 383.

' The Book of Ser Marco Poh, tranalated by Colonel H. Yule* (London, 1875),
L372.

* Tim GrUiya Sutra^j translated by H. Oldenberg, part ii^ p. 193 {Sacred Books
of the East, vol. xxx).

'' J, D< H, Temme, I>ie VoJkssatfen der Altmark (Berlin, 18tJ9), p. 73.

* Plutarch, Quaesthnes Eomanae, 29. But Catullus (ki, 166 8q.)knew better.
Compare Plautu&» Casino^ iv. 4, 1 ; Lucan, Pharsalia, ii. 359 ; J. Marquardt, Das
Frimdehen der BSmer\ p. 55; Bobinson EUia, in hia commentary on Catullus,
loc. cU,

' Varro, cited by Serrius, on Virgil, Eel viii. 29 : * Quas [soil spotisas] etiam ideo
timen ait non tangcre^ ne a sactikgio inchoarent^ si depositurae virginitatetn cakent rem
VestaCf id est numini casHsMmOf consecratam/

FOLK-LORE IN THE OLD TESTAMENT

171

modems,^ Indeed, the Romans recognized a special god of the
threshold named Limentinus, who was roughly handled by the
Christian Fathers,^ his humble station in life laying him open to
the gibes of irreverent witlings.

These facts suggest that ' the officers named Keepers of the
Threshold at the temple in Jerusalem may have been posted at the
door for the purpose of seeing that nobody committed the heinous
offence of treading on the thre^shold. For this purpose they may
even, like the warders of the threshold in the palace at Peking, have
been provided with cudgels, which they laid over the backs of all
who through ignorance, obstinacy, or accident set foot on the
sacred spot.

But while the sanctity of the threshold in many lands is ceiiain,
the reason for it is not so, and it may well be that different reasons
have been assigned for it in different places. However, there are
some grounds for thinking that the threshold has often been viewed 1
as an abode of spirits, human or othei'wise ; and such an idea would
quite suflSce to account for the superstitions which have gathered;
round it. In heathen Russia the spirits of the house are said to
have had their seat at the threshold ^ ; and consistently with this
tradition in Lithuania, * when a newly baptized child is being brought
back from church, it is customary for its father to hold it for a while
over the threshold, '* so as to place the new member of the faniUy
under the protection of the domestic divinities," , . * A man should
always cross himself when he steps over a threshold, and he ought
not, it is believed in some places, to sit down on one» Sick children,
who are suppased to have been afflicted by an evil eye, are washed
on the threshold of their cottage, in order that, with the help of the
Penates who reside there, the malady may be driven out of doors.' *
A German superstition forbids us to tread on the threshold in

' E. Tyrrel Leith, * Folk-lore of the Threshold/ Panjab Notes and Queries, I 76,
5 460 ; H. a TrurabulJ, The Threshold Cm^enant (New York, 1896), p. 36. The latter
work containB a useful collection of facta on the folk-lore of the threshold mixed up
with fiome untenable theories*

' Tertnllian, De Idohiria^ 15 ; Arnobiua^ Ad^ersus Natioms, L 28, iv. 9, 11,
and 12 ; Augustine, De Owitate Dei^ vi. 7.

' P. von Stenin, ^Ueber den Geiaterglauben in Euasland/ Globus^ Ivii (1890)^
p* 269.

* W. R. S. Ralston, The Songs qfthe Eussian People, pp. 136 sq. In SonneWrg
when a child has the cramp it is laid on the door-sill : Aug. Schleicher^ VolksiUni-
Itches aus Sontteberg (Weimar, 1858), p. 146.

172 FOLK-LORE IN THE OLD TESTAMENT

entering a new honse^ since to do so ' would hurt the poor souls ' * ;
and it is an Icelandic belief that he who sits on the threshold of
a courtyard will be attacked by spectres.^

But why should spirits be supposed to have their seat at the
threshold ? One possible answer is suggested by a Russian custom.
The peasants bury still-bom children under the threshold ^ ; hence
the souls of the dead babes may be thought to haunt the spot. But
again we may ask, Why should the bodies of still-born infants be
buried under the threshold? An answer comes from northern
. India. * When a child dies it is usually buried under the house
threshold, in the belief that as the parents tread daily over its
grave its soul will be reborn into the family/'^ A similar belief
probably explains the custom, common in Central Africa, of burying
the afterbirth at the doorway or actually under the threshold of the
hut ^ ; for the afterbirth is supposed by many peoples, for example
by the Baganda, to be a personal being, the twin brother or sister of
the infant whom it follows at a short interval into the world**^ By

* A Wuttke, Der deutsche Volksabcrglauhc \ p. 372, § 608, However, in Silesia
a contrary superstition enjoins you to be sure to tread on the threshold when you
enter a new house ; for it is thought that otherwise you will not remain in the
house a year. See P. Drechsler, SUte, Bmmh und Volksglaube in Schlemen, ii
(Leipsic, 1906), pp. 2 sq.

* F, Liebreeht, ^ur VoIksJcunde, p. 870,

^ W, K, S. Ralston, Son^s of the Busman People, p. 136.

* W, Crooke, Natives of Nmihem India (London, 1907), p. 202, A somewhat
different explanation of the custom is given by Colonel Sir R, C. Temple {Panjab
Notes and Queries, L 123, 5 925) : * A case occurred in Ambala Cantonments, in
which a humble couple, Jaisw4ras, in, for them, comfortable circumstances, were
arraigned for concealing the birth of a child* It was found buried under the
threshold. It turned out that infanticide was the last thing the parents intended,
for it was a first-bom son, and that the infant had died about nine days after birth,
and had been buried where it was found, in order that in constantly st-epping over
it the pai'ents would run no risk of losing any sul*ae<iuent children that might be
born. They said it was the custom of the caste so to bury all children that died
within fifteen days of birth/

^ Fr. Stuhlmann, Mit Emin Pascha ins Eerg von Afrika (Berlin. 1894), pp. 891,
674 ; Emm Pashu in Central Africa, being a Colledion of his Letters and Journals
(London, 1888), p. 84 j J. A. Grant, A Walk across Africa (Edinburgh and London,
1864), p, 298; J. Roscoe, * The Bdhimsi,' Journal of the Anthropological Institute,
xxxvii. 106 (1907),

" The evidence will be given in the third edition of The Goldm Bough, My
authority for the statement in the text as to the Baganda is the Rev. J. Roscoe.
For an example of a similar belief among the Toba-Bataks of Sumatra see A. G.
Kriiijt, Hd Animistne in den Indischcn Archipel (The Hague, 1906), p. 25.

folk-lore in the old testament

173

buiying the child or the afterbirth under the threshold the mother
apparently hopes that as she steps out of and into the house the spirit
of the child or of its supposed twin will pass into her womb and
be born again. On this hypothesis the widespread belief in the
reincarnation of the dead would explain the sanctity of the threshold.
But it is possible, and indeed probable, that other causes still unknown
to us have contributed to shed a glamour of mystery over that part
of the house. ^

§ 8/ The Sin of a Census.

From two well-known narratives in the Books of Samuel and
Chronicles - we learn that Jehovah cherished a singular antipathy to
the taking of a census^ which he appears to have regarded as a crime
of even deeper dye than boiling milk or jumping on a threshold.
We read that Jehovah or Satan inspired King David with the
unhappy idea of counting his people. Whatever the precise source
of the inspiration may have been, for on that point the sacred writers
differ, the result, or at least the sequel, was disastrous. The numbering
of the people was immediately followed by a great pestilence, and
popular opinion viewed the calamity as a righteous retribution for
the sin of the census. The excited imagination of the plague-stricken
people even beheld in the clouds the figure of the Destroying Angel
with his sword stretched out over Jerusalem,^ just as in the Great
Plague of London, if we may trust Defoe, a crowd in the street
fancied they saw the same dreadful apparition hovering in the air.^
It was not till the contrite king had confessed his sin and oflFered
sacrifice to appease the angr)^ Deity that the Angel of Death put
up his sword and the mourners ceaaed to go about the streets of
Jerusalem.

^ Among tliat peculiar people^ the Kafirs of the Hindoo Cooah^ the rule never to
tread on a threshold appears to be reversed : ' For some reason or other, no Kafir
seems to be able to step sedately over the raised threshold of a door. He must
spring on to it with one foot, however low the doorway, and however much he has
to bend his head. Consequently he retires in a sort of miniature whirlwind, his
loose garments floating behind him * (Sir G. S, Robertson. The Kafirs oftJie Hindu
Kush (London, 1896), p. 115). This apparently was the sort of practice which
excited the wrath of Jehovah*

* 2 Samuel xxiv ; 1 Chronicles xxu
' 1 Chronicles xxL 16.

* Daniel Defoe, History of the Plague in London (Edinburgh, 1810), pp. 33 sq.
But Defoe probably copied the narrative in Chronicles.

174

FOLK-LORE IN THE OLD TESTAMENT

Though we may not presume to fathom the grounds for this
divine dislike of a census, we can at least show that it has been
shared by savages. The Gallas of East Africa think that to count
cattle is an evil omen, and that it impedes the increase of the herd,^
And the Lapps used to be, and perhaps still are, unwiUing to count
themselves and to declare the number, because they feared that such
a reckoning would both forebode and entail a great mortality among
their people.- A precisely similar belief seems to have been held by
the Hebrews in the time of David, and the pestilence which imme-
diately followed the census was doubtless regarded by them, as in
a similar case it would have been regarded by the Lapps, as a proof
sufficient to confute the doubts of the blindest and most obstinate
sceptic.

* Ph. Paulitschke, Eiknographie Nordost-AfrikaSi die gcidige CkUtur der Dandkil,
Gulla und Somd (Berlin, 1896), p* 81.

' C, Leemitis, Be Lapponibus Fimnardiiae eorumque lingua, viia^ et religione
pristina Commentatio (Copenhagen^ 1767), p. 499 : * Censum capitum nee facUe inire
vohhantf nee prodere, meiuenteSf ne hujusmodi cmnptitatio ingctttem snorttm stragem ci
Junera portenderet et secum irahereC

THE RELIGION OF THE TORRES STRAITS

ISLANDERS

By Alfred C. Haddon

The Torres Straits Islands roughly fall into three gioups : the
western, composed of ancient igneous rocks which support a some-
what sparse vegetation; the central, whicii are mainly vegetated
banks of coral sand j and the eaatern, composed of modern volcanic
rocks and possessing a fertile soil and usually an abundance of food,
but, even here, insufficient rain during the north-west monsoon
results in a scarcity of garden produce that occasionally leads to
a partial famine.

The islanders are typical Western Papuans so far as their
physical cliaiacters, temperament, and culture are concerned. In
his recently published monograph of the languages of Torres Straits
Mr. Ray has demonstrated that the morphological linguistic affinities
of the Western Islanders are Australian, while those of the Eastern
Islanders are Papuan, and 'there is no genealogical connexion
between the two languages of the Straits'. With regard to the
fomier he says, 'It is difficult to reconcile the non-Austrahan
physical appearance of the Western Islanders with the Australian
form of their language. It has probably resulted from a gradual
occupation of the islands by natives from the New Guinea mainland.
This has gradually brought about a change in the physical features
of the people without materially affecting their language/

Practices that we tei-m ' magical ' and others that can only be
described as * religious ' entered into the daily life of the islanders
at all points/ The former were either performed by groups of men,
in which case their object was almost invariably beneficial to the
community^ or they were performed by individuals, usually for
personal ends which were sometimes anti-social. That complex of
rituals, beUefs, ideals, and sensations which we understand by the
term religion seems to be intin^ately connected with a belief in

' The detailed account of the facts given in this paper wiU be found in vols.
V and vi of the Ikports qfHie Cambridge AfMrqpological Ej^cdUion to Torres Straits,

176

THE RELIGION OF THE

a power or powers of a more or less spiritual or extra-human nature,
which can be induced by various means to assist man in securing
what he desires.

Ordinary magical practices are beyond the scope of the present
essay, but there were certain actions which seem to bridge over the
distinction between magic and religion. There was a large vaiiety
of natural and worked stones, and carved wooden figures which were
supposed to ensure good crops, and to influence animals and the
elements. Precise information is unfortunately lacking as to how
far these objects were regarded as animated. In Mabuiag, however,
wooden human effigies termed maduh were said to become animated at
night and to go round the gardens swinging buU-roarers to make plants
grow. Some had merely a generic name, while othei-s had personal
names. In most cases (possibly in all) ' sacred words ' were uttered
when the object was put into definite requisition. As in ritual songs,
the words of the sacred formula or invocation were few in number and
suggestive rather than fully descriptive sentences. Thus, the sacred
men of the enaw-fruit shrine in the Murray Islands said ^ Enau enau
turn round, stalk pluck, branches dead '. Not only were there in
the Murray Islands a large number of these objects which belonged
to the head-men of different families or to the heirs of certain
localities, but there were cei*tain sacred stones or shrines, or rituals
such as rain-making, which belonged to larger groups. When
making rain by means of liis d^iiom (a stone image of a man), a
sacred man of the rain zogo (to use the term employed by the Murray
Islanders for this class of objects), invoked various kinds of clouds,
*dark clouds, stratus clouds, overcast clouds, and rain clouds,' and
enjoined them to collect and gather, and the * noisy wind' was
called upon to * break the coco-nut leaves \ But at the annual rain-
making ceremony the following prayer was chanted repeatedly in
a small squeaking voice : ' Bain my zogo^ Give life to me (or * save
me'), And strengthen me.' The former was an invocation or com-
mand, the latter was a true prayer. On one of the Murray Islands
is a small, practically formless, stone which represents a man and
is called Waipem. In Januaiy the sacred men of this particular
shrine made an offering of fruit, and * man think inside himself '
(as the natives expressed it in jargon English), ' If we give you plenty
fmit, I think you give us plenty turtle/ They then went to the
two points of the islet to look out for the turtles which would be
sure to come. In these two instances the object prayed to liad

TORRES STRAITS ISLANDERS

177

semblance of a man, but that was not the case in the divining
oracle of Tomog zogoj to which reference will shortly be made.

Masked dances for the purpose of increasing the food supply
were very common throughout these islands. Some of these, such
as the saw-fish dance, were of a * magical ' nature, especially when
animal-masks were worn. When, however, the masks represented
a human face, we may suspect a personal element coming in ; such
were the nmwa masks of the western islands and the dogai masks of
the Murray Islands, In the latter case the ceremonies might be
termed ahnost reli^ous.

Also of a transitional character was the totemism of the
Western Islanders, which had all the essential traits that characterize
AustraUan totemism. As a social institution it was a distinct
ameliorating influence in social intercourse and tended to minimize
antagonism between communities. Individuals identified themselves
with their totem {aagzul) by decorating themselves or their belong-
ings with representations of the totem, A psychical affinity was
supposed to exist between the totem and its human kin ; thus the
crocodile-men were said to be very strong and to have no pity. The
cassowary-men were fond of fighting, and it may be noted that the
cassowary is of very uncertain temper and can kick with extreme
violence ; the members of the clan were said to be especially fast
nmners and prided themselves on their thin legs, which they likened
to those of a cassowary. If there was going to be a fight, a casso-
wary-man would say to liimself, * My legs are long and thin, I can run
and not feel tired ; my legs will go quickly and the grass will not
entangle them/ The snake-men were said to be fond of fighting,
and whenever a scrimmage occurred they got out their stone-headed
clubs and hit people, putting out their tongues at the same time and
wagging them as does a snake. The dog-men were said to be some-
times fierce and fond of fighting like the snake-men ; at other times
they were friendly and * glad to see people \ The shark-men were
also quarrelsomep On the other hand, the members of the shovel-
nosed skate clan were a quiet and peaceable folk, not given to much
talking. The ray and sucker-fish were also peaceable clans.

No man might under ordinary circumstances kill or injure his
totem ; if he did hia fellow clansmen would kill or injure him, but
any one might kill the totem of a clan other than his own with
impunity, but the clansmen of that totem would feel sorry.

There is a little evidence that the totem was invoked when

w« N

178

THE KELIGION OF THE

assistance was required. When a snake-man fought he would cry
out ' Snake bites ! ' which seems to have been a recognized formula.
The ray-men had a similar invocation. The praying to heroes,
under the appellation of augud^ belongs rather to hero-cult than
totemism.

In Mabuiag the dugong-men performed a ceremony to constrain
dugong to come towards the island to be caught, and the turtle-men
performed a ceremony to ensure a good turtling season.

It is a noteworthy fact that totemism does not occur in the
Murray Islands, and doubtless it was absent throughout the eastern
group of islands. The only probable rehcs of totemism are : (1) the
hereditary nature of certain logos^ or shrines, with which are asso-
ciated definite performances, the presumption being that originally
these were rituals connected with the increase or control of totems
by the elders of the respective clans. That these should be asso-
ciated with villages or places, quite as much as with families, is only
in accord with the assumed replacement of totemism by village
exogamy in the island. (2) The existence of a few groups of men
with animal names, which at first sight look like totemic clans. So
far as could be discovered these groups were concerned solely with
certain dances coimected with a hero-cult, which, as will be men-
tioned later, was introduced from the Western Islands. (3) The
belief in ghost-animals. The ghost of one about to die or of a
recently deceased person usually appeared to the living in the form
of some animal A kingfisher may appear for any one, but there
are certain animals that appear at the death of members of par-
ticular groups of individuals, the idea evidently being that the ghost
of a person takes the form of an animal to which it is akin, and in
that guise appears to the survivors. Usually it is the eponymous
animal of a group with an animal name that appears on the death
of a male member. Women are represented by flying animals,
bats and birds, but no relation was indicated between groups of
women and particular birds. This looks suspiciously like what has
been termed a * sex-totem ' ; but I am not prepared to admit that
these birds are totems. The ghostranimals certainly look Uke
vestiges of totemism, although there is no evidence to show that
these animals were ever connected with the social organization.

Whilst totemism as a social institution has many good points,
it seems to be too indefinite and impersonal to constitute a satis-
factory basis for an effective religion. There cannot be much

TOKRES STRAITS ISLANDERS

179

satisfaction in an indefinable alliance with a group of animals or
plants, nor can these be regarded as effectual helpers in times of
difficulty and danger. Hence it was not surprising that there was
no evidence in Torres Straits that totemism was developing into
anything more distinctively i*eligious. The need for guidance or
assistance had to be supplied from other sources.

In the Murray Islands numerous omen birds provided a means
by which men could obtain information on matters that were
beyond their ken, and there was also an elaborate oracle, Tomog zogo^
which was consulted only by the shark-men. The oracle was thus
addressed J * Tomog, you make me to know all things/ and a definite
request for information was made, to which the reply usually con-
sisted in the appearance and movements of certain animals.

Various forms of divination were constantly employed in Torres
Straits, some of which were purely * magical ', while others had
reference to spiritual influence ; of the latter by far the most
important were various methods of skull divination. For this
purpose skulls of relatives were usually employed, but this was not
essential. According to one method, the inquirer would enjoin the
skull to speak the truth, and, putting it by his pillow at night time,
would go to sleep ; the dreams were the messages from the spirit of
the deceased, and upon these action would be taken. This simple
form of divination could be performed by any person.

In consonance with the increased specialization of religious
function among the Murray Islanders, we find that skull divination
for theft was in the hands of the shark-men, who were the most
important members of the local hero-cult The diviners went into
the sacred house, taking with them the skull of a former sacred man.
One, who had to be a leading man of the group, put on a mask and
repeated certain words. They went into the bush and were led to
the house of the thief by the noise of a stridulating nocturnal insect.
A similar method was employed to discover one who had successfully
charmed or poisoned another.

The Western Islanders appear to make a distinction between
the ghost {mart) of a recently deceased person and its later stage as
a definite spirit {nmrkai) ^ ; but I have no evidence that this was done
by the Murray Islanders. If we accept the western view, we may

^ Perhaps the Eastern word mar (which has the same signifi canoe as the
western marij 'ghost, soul, reflection, shadow,' designates the earlier stage of
disemhodiinent, and the word lamar^ ' ghost or spirit * (probably an abbreviatioQ

Ii2

180

THE RELIGION OF THE

assert that the Torres Straits Islanders feared the ghost but believed
in the friendly disposition of the spirits of the departed. In Mabuiag
the corpse was carried out of the camp feet foremost, or else the
ghost would find its way back and trouble the survivors, and the food
and water of which the deceased had been partaking was placed on
or near the platform on which the corpse was laid, otherwise the
ghost came back for them and would thus annoy and frighten the
relatives. If the food was found scattered the next morning, the
people said the ghost wajs angry and threw the food about. The
ghost of a recently deceased person is particularly to be feared in
Murray Island, as it haunts the neighbourhood for two or three
months. We have native testimony for tracing the origin of the
elaborate funeral ceremonies to this belief. The Murray Islanders
perform as many as possible of the necessary ceremonies in order
that the ghost of the deceased might not feel sUghted, for otherwise
it was sure to bring trouble on the relatives by causing strong winds
to destroy their gardens and to break down their houses. These cere-
monies, some of which might take place months after the last death, or
even annually, seem to consist in Murray Island of two main elements,
(1) the dramatization of a legend accounting for various practices
connected with funerals and the journeying of ghosts to the mythical
island of Boigu in the west ; in this ceremony the chief performer,
who represented the spirit messenger, was supposed to take away the
ghosts of the dead to Boigu, (2) The pantomimic representation of
recently deceased persons in their character of denizens of the spirit
world. We are informed that the illusion of the personification of
ghosts by men was almost perfect, more especially as it is assisted
by the implicit belief of the women and children that the performei-s
really are ghosts or spirits. There is no doubt that the latter
ceremonies comforted the mourners ; probably the mam reason was
tliat it reassui-ed them that the ghostly relatives would no longer
haunt the livmg, but apart from this there may have been a real
pleasure in the idea of the return of the ghost, for we must not
forget that these aflFectionate people kept their dead in remembrance
as far as possible with their limited resources. This may be the
explanation of the stones, painted with human faces, which were
placed along a funeral screen in the island of Nagir, and to which,

of kJa mar^ ^maa's spirit*), aigoifiea the lat«r stage ; in which casd it would l>e
strictly i^quiyalent to markaiy which Mr. Eay thinks is derived from mart kai,
L e. kajsi, 'spirit person * {kazi is usually abbreviated into ka in compound. words).

TORRES STRAITS ISLANDERS

181

according to Macgillivray, were 'attached names of persons who
were dead '. I was also several times assured that the preservation
of skulls of relatives in houses was due to the sentiment of affection.

The belief in continued existence after death was very real to
these people. In the Western Islands, the ghost of a recently
deceased person remained a very intangible sort of thing for some
time after death, and it was believed to reside in the mythical island
(there called Kibu) for some time before it became a proper spirit ; as
a preliminary it had to be iiit on the head, and then was instructed
in all it had to do. The spirits behaved in every way as do men. It
is evident that the preliminary funeral ceremonies related to the
ghost, while the subsequent ones were connected with its final
entrance into the spirit world.

Certain men in Mabuiag are reputed to be friends of the spirits
and to be possessed of the gift of spirit-divination. Such a man is
termed a * spirit^touching man \ The spirit and the man talk either
in a whistling manner or by ventriloquism. Other * spirits' friends *
are tlie * ghost-seeing-men ', to whom the ghosts of deceased persons
appear. They are constrained by the ghosts to go at times into the
bush, on which occasions they become possessed, or, as it was
expi^ssed, Hum cranky, come like devil [ghost] now/

In no case have I obtained in the Western Islands an indication
of anything approaching a worship of deceased pei-sons ancestral or
otherwise, with the exception of the heroes shortly to be mentioned j
neither is there any suggestion that their own ancestors have been in
any way apotheosized.

Thanks to the efforts of Mr. John Bruce a little information on
what he regards as a kind of ancestor worship has recently been
obtained from the Murray Islands, but even he found it excessively
difficult to perauade the people to tell him anything about it. He
says, * Certain septs had their own ad giz^ who are supposed to be the
founders of their respective septs, and are reverenced and no doubt
were worshipped and supplicated on behalf of the needs of the sept/
The word ad implies anything old and traditional, with the idea of a
sanctity that is associated with ancient wont ; thus certain folk-tales
are orf, or anything about which a legend is told, and all sacred and
magical stones are ad ; Mr. Bruce adds it * signifies a god ', but I
think this is too concrete an idea, Giz means origin, base, founda-
tion, or a collection. Mr. Bruce defines ad giz as * the first god, or
god of the very b^inning of things * ; perhaps ^ancient of days' might

182

THE RELIGION OF THE

be a better term* Some of the ad were warriors, others men of
peace. Each ad has still his lineal descendant in his own sept ;
although now it does not seem much of an honour, there is no doubt
that formerly these men must have held a high position in their
respective septs. The direct descendants bear the names of their
illustrious ancestors, but these are never used in addressing them.
The direct descendants are * mere men \ and Mr, Bruce considers
* that they could not have looked on their ad as supernatural beings *,
The ad giz point back to a time prior to the introduction of the hero-
cult, and were probably the head-men of their respective groups, who
seem to have been reverenced, though to what extent is very difficult
to determine.

The exact relation of moraUty to religion is generally regarded
as a moot point, but the evidence on this point is sufficiently clear
in the Torres Straits Islands, Rules of conduct were exactly defined
and as far as possible enforced, not by officials of religion, nor by
a special judiciaiy or executive body, but by public opinion. Ulti-
mately recourse might be had to the services of a sorcerer, or to
physical force put into operation by the old men, by their delegates,
or by the friends of the injured party ; but these were merely the
recognized means by which public opinion maintained its authority
when the kno^vn disapprobation that anti-social conduct would entail
was ineffectual

Instruction in ethics and conduct was given more particularly
during the initiation period. The injunctions were : — Remembrance
of admonitions, reticence, thoughtfulness, respectful behaviour,
prompt obedience, generosity, diligence, kindness to parents and
other relatives in deed and word, truthfulness, helpfulness, manh-
ness, discretion in dealing with women, quiet temper. Bravery,
ferocity, endurance of pain and hardship and other warlike quaUties
were regarded as great virtues* The prohibitions were against theft,
borrowing without leave, shirking duty, talkativeness, abusive
language, talking scandal, marriage with certain individuals,
reveaUng the sacred secrets.

It is still believed by the Murray Islanders that a spirit may feel
resentment when childi-en of the deceased are neglected or wronged,
or when land or chattels of the deceased are taken by those who have
no claim to them. No doubt in the past such fear of the spirit's
wrath had a deterrent effect on wrong-doers and helped to keep the
people straight, ^dthough now they look rather to the civil laws than

TORKES STRAITS ISLANDERS

183

to the spirits as a means of getting their rights and punishing
offenders. With this exception, there is no evidence that their
code of morality derived either sanction or support from rehgion.
No appeal was made to totem, ancestor, or hero, and no punish-
ment from these quarters was made for infringement of social
moraUty.

So far we have briefly considered those reUgious institutions
which we may safely regard as belonging, on the whole, to the earlier
stages of the culture history of the natives of Torres Straits ; but the
most important development that took place was definitely due to
influences from the outside and resulted in the establishing of hero-
cults.

In the Westeni Islands we meet with traditions of many heroes,
some of whom were described as men ; others were relegated to a
long time ago, while some were spoken of as ad or arfi, which,
according to Mr. Ray, now signifies a legend or honorific title ; but
probably it had the same significance as the term ad of the Eastern
Islanders (see p* 181) ; to some were applied the same name, augnd^
as that by which a totem was called. Concerning those of the first
group there is told nothing, or very little, that is miraculous, they
being simple warriors or people who had adventures. Not a few
were culture heroes who introduced improved methods of agriculture
or fishings and it is in this group that the marvellous b^ins to
appear. Several introduced new ceremonies and instructed the
people in the appropriate dances. Finally a few heroes were the
objects of a special cult,

Sida^ the great culture hero for vegetable food, came from
New Guinea, where he returned after visiting the Western and
Eastern Islands of Torres Straits. Everywhere he is regarded as
a benefactor ; he instructed the people in language, he stocked reefs
with the valuable cone shell, and notably he introduced plants
useful to man. He was a very amatory person, and valuable
economic plants sprang up as the result of his amours, an example
of the close association in the native mind of the sexual act with
agricultural fertiUty, The superior fertility of Mer is also accounted
for by the introduction of garden plants from Badu and Moa by two
heroes, and at the same time this accounts for the impoverishment
of these two Western Islands, The death dances were introduced
into the Western Islands by two culture heroes from New Guinea,
one of whom brought over some funeral dances to Waier, the

184

THE RELIGION OF THE

smallest of the three Murray Islands, Two culture heroes of Mer
are reputed to have been the first to build the large weirs for
catching fish which they also introduced in some of the central
islands* At each place which they visited after leaving Mer, they
either taught a new language or suggested a different way of
speaking the old. This is the only instance known to me of
culture spreading westwards ftt>m the Eastern Islands,

The most prominent of the heroes were Kwoiam, who was
almost certainly of Australian descentj and a family of brothers who
seem to have come from New Guinea.

Kwoiam, the warrior hero of Mabuiag, made two crescentic
objects of turtle-shell, which blazed with light wlien he wore them
at night-time, and he nourished them with the savour of cooked
fish. These objects were termed aitgud^ which is the same name
by which a totem was called (presumably because the natives did
not know by what other sacred name to call them), and they became
the insignia of the two pliratries into which the old totem clans of
Mabuiag were grouped. In this island Kwoiam was designated as
adi^ and occasionally he himself was spoken of as avgiidy^hMi no
record of him or of his emblems occurs as a totem in the genealogies
of the people of Mabuiag collected by Dr. Rivers. lo the Muralug
group of islands he was regarded as the * big augud *, and the * augud
of every one in the island \ Connected with the cult of Kwoiam
were two heaps of shells, called navels of the augud^ which were
constructed to show that the two a^giid-emblems originated there,
and when it was deemed necessary to fortify the latter they were
placed upon their respective navel-shrines. The cult of Kwoiam
was associated with warfare, and when attacking an enemy the
warriors formed into two columns, each of which was led by
a head-man who wore the Kwoiam emblems. The moral value of
the augud-emhlemB in war must have been very great, and the natives
themselves recognized the fact ; as one man said : ^ Suppose we
have not an augud^ how can we fight ? ' It is recorded that on
one occasion the victorious Mabuiag men refused to continue %hting
the Moa men on account of the temporary absence of the two augtid-
men. The Moa men also had magical emblems associated with
Kwoiam, but these were not effective as compared with the former,
* because Kwoiam belonged to Mabuiag and not to Moa/

As Kwoiam was an inspiring feature in the life of the
inhabitants of the more westerly islands, so a group of hero

TORKES STRAITS ISLANDERS

185

brethren played a similar part for the natives of the central and
eastern islands.

Sigai and Maiau were the two brothers who went to Yam, and
each became associated, in his animal form^ with one of the two
pre-existing phratries. A shrine was erected for each, the essential
feature of which was a turtle-shell model representing respectively
a hannmer-headed shai^k and a crocodile, under each of these was
a stone in which the spirit, the soK^alled augnd^ resided. Outside
the enclosure which screened the shrines from profane gaze were
two heaps of shells which, as in the cult of Kwoiam, had a mystical
connexion with the shrines and were similarly termed ' navels of
the au4)ud \ The shrines were so sacred that no uninitiated persons
might visit them, nor did they know wliat they contained ; they
were aware of Sigai and Maiau, but they did not know that the
former was a hammer-headed shark and the latter a crocodile ; this
mystery was too sacred to be imparted to iminitiates. When the
heroes were addressed it was always by their human names, and not
by their animal or totem names. Warriors would be enabled to go
whither they liked if they sung certain songs at the shrines. They
prayed as follows before going to attack an enemy : * O Augiii Sigai
and Augud Maiau, both of you close the eyes of those men so that
they cannot see us.' I was also informed that when the Yam
warriors were fighting they would also call on the name of Kwoiam,
and even on that of Yadzebub, although the latter was a local
warrior who was always described as * a man *. From the folk-tales
it appears that Sigai and Maiau are more mythical or mysterious
than Kwoiam ; we have thus an instructive series of helpers :
Yadzebub, the famous local warrior ; Kwoiam, the warrior hero of
another island ; and Sigai and Maiau, the immigrant heroes whose
cult was materialized in turtle-shell images, and the spirit of each of
whom resided in a particular stone.

The account of the introduction of the hero-cult into the
Murray Islands is very perplexing. It appears that Bomai, who
was often spoken of as Malu, came first and was recognized as
a jsogo^ that is something sacred ; he was represented by a human-
face mask. Later Malu arrived with a fleet of canoes from various
western islands in search of Bomai. Malu also became a sogo, and
was represented by a mask in the form of a hammer-headed shark.
The foreigners exhibited certain dances which they gave to their
hosts and then returned home. These are the dances which were

186

THE RELIGION OF THE

referred to when dealing with the possible vestiges of totemism in
the Murray Islands. The Bomai-Malu cult predominated in the
Murray Islands, and the sacred men in connexion with it attained
considerable power wliich they frequently used for their private
ends.

At present we cannot understand the full significance of the
hero-cults that spread over the islands, for this we must await
further investigations in the district of Cape York on the one hand,
and in the adjacent region of New Guinea on the other. The
ultimate development of the cults in the islands is not without
interest. The Western Islands are not particularly fertile, so the
natives spend a good deal of their time in fishing, and there is
considerable intercourse between various islands, due to trade or
warfare. Here the hero-cults developed into war-cults. The
Murray Islands in the east are fertile, and the people are much given
to agriculture ; they are so far from other islands, except a few
insignificant vegetated sand-banks of the central group, that there
was little intercourse with other people, nor did we obtain records
of any inter-insular fighting, thus there was no inducement for the
hero-cult to develop into a war-cult, but it concerned itself more
with the social life of the people, and the three sacred men were
on the way to become priests.

Totemism was still in force in the Western Islands at the time
of the arrival of the hero-cults, but it had probably already dis-
appeared in the Murray Islands. Everywhere, but perhaps more
particularly in the Eastern Islands, there were numerous small family
or local rituals, most of which were associated with improving the food
supply. A reUgion then appeared that replaced in the west the
indefinite communal association of a totem with its clan for
a definite personal relation with a superhuman being, and it is no
wonder that it spread, being carried from island to island. These
cults also provided in both the Western and the Eastern Islands
a synthesis which had hitherto been lacking, as the men could now
meet as members of a common brotherhood, wliich was impossible
under the earlier conditions, and a feeling of intense pride in the
new cults was engendered.

An interesting parallel to these hero<!ults of Torres Straits
occurred also in Fiji The people of Viti-Levu are divided into two
groups, the Kai Veisina and the Kai Eukuruku, that trace their
descent from Veisina and Eukuruku, who drifted across the Big

TORRES STRAITS ISLANDERS

187

Ocean and taught to the people the cult associated with the large
stone enclosures, Nanga. Veisina arrived first, and where he landed
the turmeric plant sprang up> and where Rukuiiiku first placed his
foot the candle-nut grew. Their followers paint themselves respec-
tively with the yellow or black pigments obtained from these plants.
When they landed they said, ' Let us go to the Chief of Vitongo
and ask him to divide his men between us that we may teach them
tlie Nanga, for which purpose we have come to Fiji/ ^ The last
sentence points to a definite propaganda, and one is tempted to
suggest that a similar movement may also have taken place in Torres
Straits since there is not the slightest trace in tradition or elsewhere
of secular aggression.

A view has been recently expressed by L, Frobenius ^ and
W. Foy ' that some of the hero tales from Torres Straits are nature
myths. The amorous Sida, the bestower of vegetable food, is
interpreted as a sun-god, and the berserker Kwoiam as a moon-
god. I propose, elsewhere, to discuss this recrudescence of a method
of reasoning that was formerly so beloved by a certain school of
students who interested themselves in Indo-European mythology.

The tales in question must either have had a local origin or
have been transmitted from elsewhere. The theory requires that,
if the tales are indigenous, the natives must regard them as
nature myths ; but for this there is not a shred of evidence. The
burden of proof rests with the proposers of the theory if it be
contended that these are travelled tales. It is not sufficient to find
parallels in remote places for incidents in our tales, as such concep-
tions are world-wide among people in an analogous stage of culture.
Geographical continuity must be demonstrated between the Torres
Straits tales and the tales of the parent country, and, even should
the latter be proved to be intentional nature myths, it does not
follow that they were nature myths by the time they reached Torres
Straits. The Torres Straits Islanders have their nature tales, but
there is no evidence that they regard the tales indicated by these
students in that light, and tlierefore they are not nature myths for

' * The Nanga of Viti-Levii/ hy A. B, Joake, Intemat, Arck /. Ethnogr.^ II,
1889, p, 268.

' L, Frobenius, *Die Weltanschauimg der Naturviilker,* Beitr. jf. Yolks- und
Volkerkunde. Bd. VI, 1898;— Das ZeitcUter dcs SmnengoUcs, Bd. I, 1904, p. 62,
p. 189.

* W. Foy, Archwfur Edigiomwissenschaji, Bd. X, 1907, p. 129.

188 BEUGION OF TORRES STRAITS ISLANDERS

them. Why should we not regard these tales as raw material out
of which mythology may arise? According to this view a mythology
is arising in Melanesia and has arisen in Polynesia.

A review of the evidence makes it clear that the hero-cults were
not an evolution from totemism ; a transformation of totemism has
certainly occurred, but it does not appear to me to be a gradual
growth — ^a metamorphosis in the natural histoiy sense of the term —
so much as an actual grafting of a new cult upon an old. Neither is
there any suspicion that the heroes of the cults are locally-developed
ancestors, though they may have had this origin in their original
country. Unless the heroes of the cults be regarded as gods, I think
it can be definitely stated that the Torres Straits Islanders had no
deities, and certainly they had no conception of a Supreme God.

CONCERNING THE RITE AT THE TEMPLE

OF MYLITTA

By E. SIDNEY HAKTLAND, F.S.A.

Among the religious rites of antiquity there was none more alien
to modern feeling than the sacrifice of chastity by every Babylonian
woman at the temple of Mylitta. It is described first and in most
detail by Herodotus, whose denunciation of it shows that to the
Greeks of the fifth century ac, it was as abhorrent as it is to us.
According to this account, every woman once in her life was required
to sit down in the precincts of the temple of Mylitta wearing a wreath
of cord about her head, and there to wait until a stranger should
throw a silver coin into her lap and summon her with ritual words
in the name of the goddess to follow him. She was not allowed to
refuse, but was compelled to follow the first man who threw, and to
have sexual intercourse with him outside the temple. She might
then depart to her home, her duty to the goddess being fulfilled**
The historian lets fall the observation that there was a similar
custom in some places in Cyprus. This has been supposed to be
referred to by Justin, who wrote probably after the estabUshment
of Christianity, but whose work consists of selections from Trogus
Pompeius, a lost writer of the Augustan age* He reports that it
was the Cypriote custom to send maidens before their marriage on
certain days to seek their dowry by prostitution on the seashore,
and to pay the offerings to Venus for their future chastity. Dido
on her way to Carthage touched at the island at the very time, and
took on board her fleet eighty of these damsels, to be wives to her
followers and assist in peopling the city she was going to found, ^
We shall further consider Justin's statement hereafter* For the
moment we pass on to Heliopolis (Baalbec) where, the ecclesiastical
historian Socrates affirms, virgins were offered in prostitution to
strangers.^ He does not, any more than Justin, connect this with

* Herod, i. 199; Strabo, xvi. 1, 20. Further detaila are aiipplted by the
Epistle of Jeremy appended to the Apocryphal Book of Baruch.

' JuBtixi^ xvilL 5.

Socrates, Hist. Eccl i. 18.

190

CONCERNING THE RITE

a temple or a divinity ; but from Sozomen we gather that it was
a religious observance, inasmuch as the prostitution of virgins prior
to their marriage is stated to have been abolished by Constantino
when he destroyed the temple of Aphrodite.^ A similar custom,
according to Aelian, was followed by the Lydians. And he expressly
says that when once the rite had been performed the woman
remained ever afterwards chaste, nor would a repetition be forgiven
her on any plea.- Herodotus, however, states that the daughters
of the common people in Lydia earned their dowries by a life of
prostitution*^ The two writers are obviously referring to two different
customs. A third custom distinct fi'om either is mentioned by
Strabo as practised by the Armenians, among whom even the
highest families of the nation consecrated their virgin daughters
to the service of the goddess Anaitis, to remain as prostitutes at her
temple before their marriage,*

What is the relation of these three customs? They have
usually been considered as closely connected. It may be, as
Dr. Frazer suspects, that the real motive for the custom described
as that of earning dowries by prostitution was religious, rather than
economical, although my own suspicions point in another direction.
But putting that custom aside for further examination, both the
others are certainly portrayed as reUgious. As practised by the
nations of Western Asia for a thousand years prior to the fall of
paganism they were annexed to the cult of certain divinities. There
is, however, a broad distinction to be drawn between a custom
requiring every woman once in her lifetime to submit to the
embraces of a stranger, and one which consecrated a life of prostitu-
tion. Such a life was one of devotion to the goddess as a more or
less permanent servant. The other custom demandod a single act
which freed the wordiipper for the rest of her days. It may be
freely conceded that the goddess at whose temple, or on whose
feast-day, the act was performed was endowed with similar character-
istics to those of the goddess in whose service the life of prostitution
was Uved. It may even have been that sometimes the same goddess
had bands of harlots attached to her shrine, and also i^uired the
sacrifice of the virginity of all other women in the manner described.
This perhaps, as we shall see, was the case in Lydia. We should

^ Sozomenj Hist. Eccl v. 10.
^ Herod, i. 98,

' Aelian, Var, Hist iv. 1.
* Strabo, xi. 14, le.

AT THE TEMPLE OF MYLITTA

191

still need to investigate separately the two customs. One of the
most fertile sources of error in the interpretation of custom is the
fatal tendency of rites distinct, or even altogether diflferent in origin
and intention, but similar in expression, to converge* This con-
vergence is accelerated by a variety of causes. The natural vague-
ness of tradition, the forgetfiilness of the exact original meaning,
the gradual predominance of one idea over another owing to
circumstances which, for want of knowledge, we call accidental, the
tendency to repeat by way of precaution in one rite acts which
eeeentially belong to another, are all causes of the kind referred to.
Moreover, we have so often found in the similarity of rite the real
key to a common interpretation, that where convergence does not in
fact occur there is a temptation to read identity of meaning into
two rites having a superficial likeness. It behoves us, therefore, to
be on our guard, and to scrutinize with some scepticism all cases
where the identity both of act and intention is not demonstrably
complete.

The practices I have enumerated have all been interpreted as
expiations for marriage- Marriage, it is said, — the appropriation of
one or more women to one man— is an evolution from the primeval
condition of promiscuity. Religioiis prostitution, the jtis primae noctis
and other customs are expiations exacted by society from women
who are thus appropriated. They witness to the primeval common
rights of the male sex, thus asserted for the last time by one or
more on behalf of all on abandoning the woman to the exclusive
possession of one of their number.

Now, if the interpretation in question be suitable for any of these
customs, it is more suitable for the single rite such as that at the
temple of Myhtta than for the exercise of prostitution over an
extended period ; and it is to this rite that I desire more particularly
to call attention in the present paper, I need hardly observe that
the explanation of the rite as an expiation for marriage does not
by any means follow of necessity from the theory of primitive
promiscuity. On the contrary it overlooks one of the peculiar
features of the rite. Alike at Babylon, at Heliopolis, and apparently
at Cyprus (if Cyprus be a case in point} the act has to be accomplished
with a stranger. If it were a forfeit rendered to the general body
of men, who might have had a claim to temporary union but for the
institution of marriage, or if it were a formal witness of that claim,
it would seem, prima facie, more natural that it should be accom-

192

CONCERNING THE KITE

plished with some or more of the claimants, that is to say with
a member or members of the same community, A similar rite of
intercourse with a stranger was practised, as Lucian relates, at
Byblus. There it was the custom at the mourning for Adonis to
perform the well-known mourning rite of cutting off the hair. Any
woman wlio refused to do this was required to exhibit herself on
one day of the festival and undergo prostitution to one of the
strangers who resorted tliither, handing over the price to the goddess
called by Lucian the Byblian Aphrodite,^ The rite as there practised
therefore was, at all events in the second century a, d., an alternative
to the dedication of hair : it was a redemption for the tresses that
should have been sacrificed. Thus the woman would repeat the
expiation once a yeai% whether married or single, so long as she
was unwilling to shear her locks, or preferred the alternative
sacrifice of her chastity. Tliere is no evidence that it ever had
anything to do with marriage ; it certainly had not when Lucian
wrote.

The rite at Byblus must, however, be distinguished from those we
are considering. They were performed by every woman without
alternative, but they were performed only once. If they were an
expiation for marriage we should expect to find them described as
part of the marriage rites. The Balearic islandei-s, the Nasamonians
and the Auziles in antiquity had, as well as many modern savages,
such rites, whether or not they can be properly explained as an
expiation for marriage. But at the most the rites with which we
are now concerned were a preliminary to marriage — a necessary
preliminary, perhaps, but one that might have been accomplished at
any period before it. Indeed, so far as appears from Herodotus, the
\actim, if we may call her so, of the Babylonian rite was not
necessarily unmarried. But comparison of the accounts of the
practice at Heliopolis, in Lydia and in Cyprus rendei-s it fairly
certain that it was only unmarried women who were subjected to it,
and that it was essentially a sacrifice of maidenhood. A passing
reference by Eusebius has been interpreted to imply that at Helio-
polis both married women and girls were prostituted in the service
of the goddess.- But Eusebius says nothing about the goddess.
His reference must be constiTied in the light of Socrates* statement
that women were by the law of the country required to be common ^

Luokn, De Dm S^ia, 6,

* FrazBT, AdaniSf 22, note 2.

AT THE TEMPLE OF MYLITTA

193

and hence the oflEspring was doubtful, for there was no distinction
between fathers and children.* TATiatever else those phrases may
mean, they entirely negative the theory of expiation for marriage.
But they do not refer to the custom of prostituting vii^ins to
strangers, which the historian expressly distinguishes.

It may be objected to this reading of Herodotus that while he
uses the generic term women (yvi/atices) in speaking of the victims,
on the other hand, in a previous chapter referring to the Babylonian
marriage customs, he reports that once a year in every village the
marriageable maidens {TrapOtvoi) were all put up to auction^ the
respective purchasers being required to give security that they
would marry them ; and it was unlawful to give them in marriage
in any other way, Tlie objection is of little weight. It is needless
to consider whether we are to understand the specific term irapdivoi
literally. Even if so, there would doubtless be ample time for the
performance of the rite at the temple of Mylitta between the auction
and the marriage. It does not appear that marriage followed the
auction immediately. Had that been contemplated, security would
hardly have been necessary. When the anniversary came round all
the maidens who had during the preceding year attained puberty and
thus become ripe for marriage (ya/tcwi/ mpoXai) were probably put up.
Those who had not previously undergone the rite would, if my
interpretation be correct, be required to submit to it before marriage.

The theory of expiation for marriage has been so generally
abandoned by anthropologists that it is superfluous to discuss other
and obvious objections to it. But the appearance of prostitution
which the rite presents demands further consideration. At Babylon,
although a piece of money passed, the payment seems to have been
merely pro fon^ia. It mattered not how small the coin was, it
could not be refused. Whatever it was, Strabo tells us it was
considered as consecrated to the goddess. Lastly, the rite once
perfoimed, no gift, were it ever so great, would be accepted to repeat
it. The details of the rite at HeUopolis and among the Lydians
have not been preserved to us ; but we may with probability infer
that they were similar. In Lydia, indeed, if we are to trust both
AeUan and Herodotus, two distinct customs are traceable, namely,
the sacrifice of viiiginity and the life of prostitution to earn a dowry.
A Greek inscription of the second century a. d», found at Tralles and
referred to by Dr, Prazer, discloses also the existence of religious

* Socrates, loc. cit

rthom O

194

CONCERNING THE RITE

prostitution by girls expressly chosen by the god and set apart for
that end.* This is a similar custom to that of the Armenian girls
already mentioned, and is not to be confounded with the prostitution
mentioned by the Father of History as practised by all the daughters
of the conmion people. Whatever may have been the origin of the
latter, the other two in the time of AeUan were connected with
reUgion. On the island of Cyprus we seem to find much the same
state of things. If we may believe Justin, the maidens earned their
dowiy by prostitution. From other sources we learn that there were
mysteries of the Cypriote Aphrodite, which were said to have been
instituted by Cinyras, king of Paphos and father of Adonis. Into
these mysteries there was a regular initiation. Sexual matters no
doubt formed their staple teaching ; and what classical and especially
apologetic writers would call prostitution would be practised. The
legend ran that the daughters of Cinyras, through the wrath of
Aphrodite, united themselves with strangers.'^ Probably it was
believed to be in imitation of them that the maidens of Cyprus
sought prostitution on the sea-shore. In any case the story in-
dicates, as Dr. Eraser has pointed out, * that the princesses of Paphos
had to conform to the custom as well as the women of humble
birth.' But if this be so, the object of the harlotry alleged by Justin
falls to the ground, since it would be unnecessary for princesses to
earn their dowry. It may be suspected, therefore, that Justin or his
authority has confounded two disparate customs, that of earning the
dowry by pix>stitution, and that of a religious sacrifice of virginity in
connexion with the mysteries of Aphrodite, in which the other
party to the rite was a stranger. Only thus can we satisfactorily
explain the limitation of the practice to stated days, probably festivals
of Aphrodite, and the phrase about paying the offerings to her for
future chastity.

The money payment, whether large or small, was in the
Byblian rite, as in the Babylonian and (if I interpret correctly) in
the Cypriote rites, consecrated to the goddess. We may infer that
the same was the case wherever else the rite was performed. At
Byblus it was the alternative to the consecration of the woman's

' Bamsay, Cities and Bishoprics of Phrygia, i. 94, 116 ; Frazer, Adonis^ 23,
Such religious pi-ostitutes were, of course, common m Western Asia. C£ Strabo,
xii. 3, 86.

" Clement of Alexandria, FrotrepL ii ; Arnobius, Adv. Gtntes, v. 19 ; Ftrmicua
Materaus, De Erron Prof. Rd. x ; ApoUodorus, BlbL iii. 14, 3,

AT THE TEMPLE OF MYLITTA

195

hair. Prostitution — that is, sexual intercourse for hire — is not a
primitive practice ; it is a product of civilization. Tlie appearance
of prostitution in connexion with religion may be accounted for by
the influence upon the religious practice of the general practice of
harlots. Analogy would suggest that intercourse other than conjugal
or the satisfaction of the genuine passion of love demanded a
monetaiy consideration. But wlien that intercourse was the per-
formance of a religious duty the money was not kept as gain by the
woman. It was not earned for herself, but devoted to the goddess.
Where bands of ' hai'lots * were attached to a temple their earnings
probably went to swell the temple funds out of which they were
supported, ' It may accordingly be suggested that the hire was not
an essential part of the rite, but merely an aftergrowth in the process
of adapting an older custom to the cliangmg manners and reUgious
ideas of a growing civilization.^

Assuming, therefore, that the rite was a sacrifice of virginity
to which every woman was subjected, it would probably l^e per-
formed either on the attainment of puberty or as a preliminary to
the marriage ceremonies. But we gather from the historian's account
of the sale of the village maidens around Babylon that the auction
followed almost immediately after the attainment of puberty, or
witliin a year at the furthest. The practice of most ancient
nations, as of nearly all barbarous and savage peoples, and indeed
of many in a high stage of civilization, would lead us to
expect that marriage would be entered into within a very short
time of the bride s puberty. Sometimes marriage even precedes
puberty. Where, as more usual, it follows that epoch of life the
rites incident to puberty must first be completed. Among such
rites defloration is not infrequently found. In this respect the
Australian tribes are notorious. In the Boulia district of Northern
Queensland the girl is compelled to intercourse mth a number of
men.^ Among the Dieri of South Australia a ceremony called

' The service of the hierai is discussed by Ramsay, op. cU., 135-7. See also
below, p. 198.

' On the other hand it must not be forgotten that at a marriage among the
Auziles and the Nasamonians the guests who enjoyed the bride^s favours were
expected to reward her with a gift* Similarly, in modern Europe, a gift is also
found as the return for a kiss or a dance with the bride. I have collected several
cases in Legend of Perseus^ ii. 361, 365-8, and many more might be added.
Compare the Suahili custom mentioned below^ p. 197.

' Roth, Ethnol Studks, 174.

O 2

196

CONCERNING THE RITE

Wilpadrina is performed on the young women when they come to
maturity, in which the elder men claim and exercise a right to them,
and that in the presence of the other women. * The Arunta and
Ilpirra tribes in the centre of the continent perform a ceremony on
every girl when she arrives at a marriageable age, but before she
has been taken over by the husband to whom she has been allotted.
As part of that ceremony a number of men have access to her in
ritual order; and the intercourse is often repeated the following
day.^ Analogous proceedings are known in other parts of the
world. The central tribes of New Ireland have a women's house in
every village. When a girl attains puberty she withdraws into
a small house, called mbakf built inside it There it is said she has
to remain for ten months, only going outside at night. During this
period she is waited upon by the old women, and through their
interv^ention evei-y man who chooses has access to her. On leaving
the fubak she belongs only to the husband to whom she has probably
been betrothed since infancy.^ In the west of the island of Serang
between Celebes and New Guinea, a girl after ceremonial bathing goes
round clothed with a sarong woven of the fibre of the Pandanm repens,
at the service of every man until her family have collected the
necessary materials for a feast. In cei*tain districts^ however, before
actual puberty the teeth are filed. When this operation is completed, a
feast is prepared of which the novice must taste everything. Further,
an earthen pot filled with spring water is covered with a fresh
pisang-leaf. One of the old women then taking the index*finger
of the girl's right hand thrusts it through the leaf as * a symbol
of the rupture of the hymen, or to show that the possession of
virginity means nothing for her'. The leaf is then displayed on
the ridge of the roof. This done^ the women fall to eating and
drinking. When they have finished they begin singing to the
accompaniment of drums. The men are then admitted to the
house. In some villages the old men have free access tliat evening
to the room of the girl in whose honour the feast is given, while
the other guests amuse themselves with singing outside- After this
celebration the girl is entitled to free intercourse with men, even

' Howitt, Native Tribes, eU ; J. A, J., atx. 87.
Wcih, i. SOS.

• Spencer and Qlllen, Native TriheSt 92.

* Globm, xci, SI 3.

Bee also PI033 -Battels, Das

AT THE TEMPLE OF MYLITTA

197

before puberty J In east-central Africa the AzLmba maiden is
artificially deflowered during a period of retirement and instruction
in the forest. When the retirement is over she celebrates her
attainment of puberty by a dance in which only women take part.
That night a man, hired by her father for the purpose, sleeps with
her, and once this is done she is supposed to have no further inter-
course wth him. Often, however, she is already married before
puberty, and consequently no longer a maiden. None the less is
she taken from her husband that the puberty customs may be per-
formed. When she is brought back he himself sleeps with her
apparently as a ritual act, without the necessity of hiring a man for
the purpose. =^ Among the Wanyasa, or Mang anja, at the southern
end of Lake Nyasa, ceremonies are performed similar to those of
the Tntonjane {girls' puberty ceremonies) of South Africa, and every
girl on her return after the initiation must find some man *to
be with her', otherwise she will die,^ The Intonjane among the
Kaflirs is well known to be an occasion of sexual indulgence. It
may be surmised that the ceremonies of the Suahili on the east
coast were originally sunilar to those just mentioned. But the
SuahUi have become partially Arabized, though their Mohammedanism
is little more than a veneer over their heathen customs and belief.
Among them now a girl returns from her seclusion in silence
and gives her hand to every man she meets, receiving from him
in retmn a few small coins/ It is said that the girls of the
Wamegi, also a tribe near the coast, are artificially deflowered at
puberty by certain old women. ^ Artificial defloration at puberty
is also practised by the Sawu Islanders. The Sakalava girls in
Madagascar perform it on themselves in case their pamnts have not
previously taken the trouble.** Other examples could be cited, but
the subject need not be pursued.

I would venture to suggest then that the Babylonian rite was

' Riedel, Ik, Sluik- en Kroesharige Hmsen, 1S8, 137.

* H. Crawford Angus in Zeits. f, Ethfmhgie^ xxx. Verhandl, 479.

' DufF Macdonald, Africana, h 126 ; Jaa* Macdonald, in J, A, /., xxii. 101»

* H. Zacbe, in Zeits, / Etimol, xxxi* 76. Aboot thirty yearn ago a French
writer cited by Hertz {Dk Sage vom Giftmadclteriy 41) reported that among the
Bafiota of the Loango Coast the girls were led round the village and their virginity
put up to auction, Thia looks like a puberty rite of a similar character. I have
not Keen the book^ however^ and think it not impossible that the writer may have
misunderstood the ceremony usual on emerging from the * paint-house \

* J. A. L,xxxl 12L • PloBs-Bartels, Das Weib, I 307, 808.

198

CONCERNING THE RITE

a puberty rite, and that a maiden was not admitted to the status
and privileges of adult life until she had thus been ceremonially
deflowered. Among those privileges, and the cliief of them, was
the gratification of the sexual instinct. It was, therefore, a pre-
requisite to marriage. Ceremonial defloration of the bride by others
than her husband has prevailed in many places. When marriage
foUows closely after puberty it is difficult to determine whether the
custom really belongs to the puberty rites, or to those of marriage*
I am not concerned here to deny that among many peoplas who
practise it as part of the marriage rites it may have been such ab
initio. The determination of this question would involve an ex*
amination of marriage customs extending far beyond the space at
my disposal But it will be admitted that as puberty rites gradually
became simplified or altogether obsolete such a custom could only
maintain existence as part of the marriage rites. It is then usuaUy
performed by one or more of the bridegroom's friends or by an
appointed official, and ultimately degenerates into the jus primae
noctis vested in some powerful personage, as a lord or priest.
Nothing of the sort appears in the accounts which have come
down to us of the ancient rite in Western Asia. In all of them
(save among the Lydians) emphasis is laid on the performance by
a stranger. At Babylon, as we have seen, our information does
not connect the rite with marriage at alL Elsewhere it is referred
to not as part of the maixiage rites, but as a preliminary to
marriage.

That such a rite should be found annexed to the temple and
worship of a luxurious goddess causes no surprise ; on the contrary,
it is what might have been anticipated. Every reader will call to
mind numerous examples of archaic rites which have become attached
to Christian festivals, and of Christian shrines which are simply
shrines of an earlier religion adopted and consecrated afresh under
Christian names. The difficulty of uprooting old customs, and their
consequent incorporation and adaptation by advancing culture or
a new religion, aie phenomena too well known to be insisted on
here. It is probable that other customs, such as the prostitution of
the Armenian girls at the temple of Anaitis, or that of the Lydian
and Paphian girls to earn their dowries, are no more than the
adaptation of a custom common enough in the lower barbarism, by
which unmarried girls have unfettered Uberty in their sexual rela-
tions. The Armenian maidens, at all events, though spoken of as

AT THE TEMPLE OF MYLITTA

199

harlots by Strabo, do not seem to have exercised their calling for
money, nor to have admitted indiscriminately to their favoura all
who offered. They reserved themselves for their equals in rank,
and entertained them in theii- dwellings with more hospitality than
in a spinsters' house in the Pacific Islands. The surmise may be
indulged that it was in fact originally, if not in later times, their way
of choosing husbands. The Lydian girls are expressly said to have
bestowed themselves in marriage.

Mannhardt contended (and his opinion is so far endoi'sed by
Dr. Frazer) that the maidens who surrendered their virginity in con-
nexion with the cult of a goddess like Aphrodite did so in imitation
of their divinity, as her representatives, the human players of her
part.^ This may have been the mode by which the ancient custom
was adapted to the newer order of things. But it is submitted that
it is a very insufficient account of it. The custom must have been
older than any definite beUef in the goddess s habits or any story of
her various intrigues. Are we then to suppose that it was a magical
rite designed to promote the fertility of animal and vegetable hfe ?
Such rites are known in both hemispheres. The great goddess wor-
shipped under different names throughout Western Asia personified,
we may concede, the reproductive energies of Nature. Many of the
rites employed in her cult are in the last analysis magical, and had
for their purpose to assist those energies. By a well-known mental
process magical efficacy is often ascribed to acts and usages not
essentially of a magical, nor indeed of a ritual^ character. Thus the
general prostitution of young girls to earn their dowries, and that of
widows — customs which are probably of quite a different origin —
are among certain tribes of Morocco held to be not without their
effect on the abundance of the crops." Such a beUef may have con-
secrated lives of habitual harlotry in Armenia, in Lydia^ and in
Cyprus* It by no means foDows that every rite performed in the
name of the goddess acquired that meaning, still less that that was
its primitive meaning. Many such rites would be wholly personal.
They would be intended to secure personal blessings to the wor-
shipper, and notliing more, though every one might have been
required to perform them. It is needless to suppose without express
evidence that the rite described by Herodotus as taking place at the
temple of Mylitta had more than a personal reference.

* Manniiardt, Wald- mid Ftidkulie, ii. 284 ; Frazer, AdoniSf 21 note,
* Mev, Hi$t Rel, xli. 315.

200

CONCERNING THE RITE

The most difficult of all the problems connected with the rite is
to explain why it must be accomplished with a stranger. The diffi-
culty, however, is not peculiar to the interpretation here suggested.
If, as has been alleged, the act of defloration of a maiden were held
to be in itself dangerous, it is not easy to say why any one, even
a stranger, should undertake it, unless he were strangely ignorant of
the risk or strangely careless. In some places, indeed, a maiden
who had come to submit to the rite may have been outwardly indis-
tinguishable from one of the hierai ; and henoe the man may have
been unconscious of his risk, or may have been willing to undertake
a risk thus diminished. But at Babylon the women who came thus
to oflfer themselves wore a distinctive head-dress of cords, the emblem,
perhaps, of their condition of virginity. Moreover, they seem to have
been penned in enclosures divided from each other by ropes, which
were broken to let them out for the accomphshment of the rite.
There was therefore no mistake as to their status or object. On the
other hand, if the defloration simply involved ritual impurity such
as could be removed by the proper ceremonies, it must be asked why
the task was left to a stranger. None of our ancient authorities have
condescended to define a stranger. We are probably to understand
by that term one who was not an inhabitant of the town or who was
not a member of the community. The analogy of certain Australian
rites already referred to, and of rites of marriage in some other parts
of the world, would lead us to suppose that what was really intended
in the first instance was one who was not eUgible for sexual relations
with the woman in the oMinary course. Thus in Peru and New
Granada * the nearest relations of the bride and her most intimate
friends ' are said to have performed the corresponding rite ^ ; and
even her father is credited with the labour among the Orang-Sakai
of the Malay Peninsula, the Battas of Sumatra, the Alfoers of Celebes,
and on the island of Ceylon and the eastern Moluccas.^ From this
the more developed moraUty of the Babylonians would recoil.
Mr. Crawley, commenting on the Australian rite, surmises that in it
' initiation ' and marriage are one, and that * initiation * ceremonies
(that is to say, puberty ceremonies) ' of this kind are marriages to the
other sex in abstract \^ The surmise follows froiii his theory of the
danger of human contact, and especially of marriage, and the impor-

' Garcilasao, i 59,

■ Ploss-Bartels, cp, cit, 406 ; Hertz, loc ciL, citiDg authorities,
' Crawley, The Mystic Rose, 348.

AT THE TEMPLE OF MYLIITA

201

tance of ceremonies to avert the peril. The theory itself — at all
events pushed to the length to which Mr. Crawley pushes it — is very
questionable. But defloration at puberty, whether natural or arti-
ficial, is undoubtedly (whatever else it may be) a formal introduction
to sexual life. Such introduction might be the more authoritative
and emphatic if given by one (or more) with whom sexual relations
would not in future be sustained. It is a ritual act. Eitual acts are
acts out of the ordinary course— often clean contrary to the ordinary
course. Therein consists their essence, their virtue. But in the
growth of civilization, with the emergence of a new reUgion or
different customs, the real meaning of a traditional rite is obscured,
the rite itself becomes decadent, and a new meaning is assigned to
it* Hence a puberty rite might easily become part of the cult of
a goddess Uke Myhtta,

• At the stage of decay which the rite had reached at Babylon
and elsewhere in Western Asia, the proviso that the person with
whom the act was performed must be a stranger might be intended
to prevent an assignation. When the act had to be performed as
a sacrifice in honour of the goddess it might be regarded as a pro-
fanation to perform it as an act of inclination with a favoured lover.
The best way to prevent this would be to require that it should be
performed with a chance stranger, who might further be looked upon,
if Mannhardt's interpretation be correct, as a representative sent by
the goddess to play Adonis to the maiden's Aphrodite, The rite at
Byblus lends countenance to this conjecture. It is supported also by
the artificial defloration enacted only in symbol by Roman brides,
but in grim earnest at the temples of Siva by brides in Southern
India. From a sacrifice of this kind it is only a step to the substitu-
tion of the priest for the image of the god, and the way is opened to
the abuses of the jus primae noctis.

The conjecture thus presented is offered for what it may be
worth. It fits the interpretation of the rite as a puberty ceremony,
and raises, I think, no fresh difficulties. Whether it is a satisfactory
solution of the problem depends on the exact purport of the rite re-
garded as a preparation for sexual relations : a question demanding
far more space to investigate than can be conveniently occupied
here.

Since writing the above I have had an opportmiity of reading
Professor Cumont s note on the subject of religious prostitution in
his recent work, Les Meligions Orientales duns k Paganisnw Ramain^

202

THE EITE AT THE TEMPLE OF MYLITTA

pp, 148-286. He makes no distinction between the three customs
of sacrifice of virginity, prostitution to earn a dowry, and a Ufe of
religious prostitution in the service of the goddess. He refers them
all to the primitive constitution of the Semitic tribe, and explains
them as a modified form, become utilitarian, of an ancient exogamy.
Mating with a vii-gin, he holds, resulted in defilement ; therefore
she was given first to a stranger ; only after that could she be
manied to a man of her own race, I pass by the confusion between
the three customs in question, to all of which his explanation wiU
not equally apply. But if the explanation be correct for any of
them, either the ancient exogamy of the Semites must have been
quite different from exogamy as generally understood, or it must
have been not merely modified but transformed. Exogamy, as
generally understood, has nothing to do with race or nationality. It
is simply the savage rule corresponding to our table of prohibited
degrees J A man may not marry or have sexual relations with one
who is akin to him ; every member of Ixis clan (not of his tribe or
his race) is akin to him ; therefore, he cannot marry or have sexual
relations with any member of his clan. The origin of this rule is
still disputed by anthropologists, and we need not here discuss it.
But since exogamy bars a man from sexual relations with every mem-
ber of his kin, it is obvious that it cannot be merely a preliminary
to marriage within the kin. Where exogamy is the law, the bar is
absolute ; it is the law for the whole of life ; it is not intended to
provide for a temporary union outside the kin in order to prepare the
way for a permanent union within the kin. Exogamy, therefore, I
submit, cannot explain these customs.

^ I am reminded by the editor that among many savages additional prohibited
degrees exist eide by aide with exogamy strictly so called. In my view these,
where they exist, are supplementary rules of subsequent growth. In any case
exogamy operates in the same way as our prohibited degrees.

AUSTRALIAN PROBLEMS

By ANDREW LANG, M,A, LL.D.

The Puzzle of One Totetn to One Totem MmTiage,

Our chief authorities on Australian institutions, Messi-s, Spencer
and Gillen and Mr, Hewitt, have not, I think, offered any speculations
on one of their most remarkable discoveries* Tliis is the rule of
' one totem to one totem marriage '. As described by our authors,
this amazing law appears to be out of harmony with all our
speculative conclusions concerning the rise and development of
exogamy. For example, let the inquirer read Mr* Howitt's " Sum-
mary of Limitations*' (on marriage) in N,TS.E,A., pp. 282-6,
Space forbids me to quote the whole of these pages, but Mr. Howitt
says (p. 282) : ^ There is first of all ^ the segmentation of a whole
community into two exogamous intermarrying moieties, thus limiting
the choice of a wife to one half of the women in a tribe/ — while
pai'ents do not marry their children, nor brothers their sistei^. Next
there is the Umitation to the JVoa, Nupa^ or Unawa sets of potential
husbands and wives ; the range of potentiality varying in various
tribes. Then there are the systems of four sub-classes, and of eight
sub-classes ; the field of choice being thus progressively limited.
* When we turn to the totems, we find that there also this system of
limitations obtains, for in some tribes marriage is only permitted
between certain totems on either side, and not, as for instance in the
Dieri, between any of the totems on one side, and any of the totems
on the other This again limits the number of women otherwise
available/

An example of the limitation of certain totems on one side to
certain totems on the other is given by Mr. Howitt in the case of
the southern Urabunna (pp. 93, 187, 188),* * The table is evidently

* My italics.

* Mr. Howitt's source is Mr, J, Hogarth, of whom (if I do not misunderstand
him) he says that * my correepondent had not theii* scientific training or wide
knowledge of the subject'. *They' are Messrs. Spencer and Gillen (p. 282).
As many tribes, including the Wonkan^ni, apply the same titles, such as

204

AUSTRALIAN PROBLEMS

imperfect ', he says, butj clearly enough, some totems may marry into
two or even three totems in the opposite phratry^ and vice versa,
This limitation appears to me to be only a foi-m or gemi of the four
sub-clafis system. The Buntamurra tribe (with female descent) has
that system ; and, as the tribesman who gave information averred
that * certain totems belonged to his sub-class ', he must also have
conceived that only certain totem kins could marry into certain other
totem kins {Howitt, pp. 113, 114, 226^7). When Mr. Howitt tells
us that among the southern Urabunna (the Yendakarangu) * the rule
is that certain totems of the one class are assigned to certain totems
of the other' (p. 187), he seems to me to indicate at least the germ
of the four-class system as it is understood by his Buntamurra
informant, who, as Mr, Howitt shows, misunderstands it (p. 227),

Leaving out of view the locai limitations of the Kurnai and
others, we have now seen all the limitations presented by Mr,
Howitt— in his Summary. They have, he concludes, * the effect, no
doubt intended, of preventing marriages of pei^sons of too near flesh.
All these complicated and cumulative restrictions were certainly
made intentionally to meet a tribal sense of morality ' (p. 283).

Mr. Howitt, in his Summary, we see, has omitted the strangest
and most stringent of all reported limitations prevailing among
tribes which he regards as in the first and second grades of
primitiveness. The northern Urabunna with two phratries, no sub-
clauses, and ' group-marriage ' (so styled), also the Itchumundi and
Karamundi * nations ' (some ten tribes, cf, pp, 49, 50), and,
appai'ently, *some' of the Barkinji tribes {?),^ have two phratries,
female descent, and no sub-classes, but none the less possess the
greatest number of Umitations, and oflFer by far the most restricted
field of choice in marriage. Though they have two phratries, they
do not limit the choice of a wife to * one half of the women in a
tribe ' (p. 282). Tliey have exactly as many intermarrying exo-
gamous divisions as them are totems in the tribe (pp, 189, 194).
One totem, say Eagle Hawk in the Mukwara phratry, may only

murdu, and hamir^ to * both claaa' {phratry) * and totem' (p. 91), it has occurred
to me that the northern Urabunna informants may have been speaking, not
of toteroe, but of eub-cl asses of animal name, each including three or more
totems, A misunderstanding might thus arise between the informants and
the inquirers,

' I am not certain that Mr. Howitt intends to include any Barkinji« See
pp. 189, 194.

AUSTRALIAN PROBLEMS

205

marry into one totenij say Emu^ of the Kilt>ara phratry. For
the same rule among, at least, the northern Urabunna, see Messrs.
Spencer and Gillen, KT.CA,^ p, 60 and note 2 ; Northern Tribes C\A.^
p. 71, and Hewitt, p. 188, where Mr. Spencer tells Mr. Howitt
that the northern Urabunna ' were very empixatic ' on this rule.

' In the first place ' (that is, in the first place after the phratry
limit), say Messrs, Spencer and Gillen concerning the Urabunna
{Northern Tribes^ p. 71), *men of one totem can only many women
of another special totem/ How can we be sure that the phratry
division (Matthurie and Kirarawa) is prior in institution to the
limitation of one totem to one totem, which is placed second ? If
there were no phratry divisions at all, the results would still be as at
present, one totem can only marry into one other totem. Elsewhere
{N.T.CA.j p* 60, note 1) our authors write : * The fiindamental fact is
that men of one moiety of the tribe must marry women of the
other.' That is a feet — no one totem existing in both moieties — but
is it * the fundanienM fact ' ? Surely the fundamental fact is that
men of one totem must only marry women of one other totem.

Though they state this rule, our authors write, on the same page,
that ^nong tlie Urabunna tribe, * division lias not proceeded beyond
the formation of the two original exogamous moieties.' By their
own account, however, division has proceeded so much further that,
out of, say, sixteen totem kins, one totem kin can only marry into
one other totem kin, in the opposite phratry. There are thus, in
allf not, as among the northern Arunta, eight exogamous sets, but
sixteen exogamous sets; there are more, if there are more than sixteen
totems. Each totem is, practically, a phratry.

Now this amazing law is reported only from tribes very
* primitive *. We do not know but that it is the earhest rule of all.
In any case it is very early ; consequently some very ' primitive
tribes * outdo the most advanced eight^lass tribes in the stringency
of their regulation. The inevitable result is to promote marriages
among ' people of near flesh '. If the Gordons could only marry the
Forbeses, as Eagle Hawk can only marry Emu, then manifestly
Gordons and Forbeses would become practically of the same blood.
But they are large populous clans. How many persons are comprised
in the two exogamous and intermarrying totem kins ? Very near
kin they must all be ; so where is the sentiment of tribal morality
against their unions ?

Granting that our information about these tribes is correct, what

206

AUSTKALIAl^r PROBLEMS

becomes of all our theories about exc^aniy ? I have read no com*
ments on the facts beyond my own expressions of conjecture in TJie
Secret of the Totem^ pp, 186, 186. Take the theory of Mr. Spencer,
Mr, Howitt, Mr. Frazer, and Mr. Hartland. A previously * undivided
commune * bisects itself, for reasons variously guessed at, into two
exogamous intermarrying moieties. In each there somehow were {or
have come to be) so many distinct totem kins. Why do men next
bind pairs of these kins to exogamy and intermarriage with each
other alone ? If that rule, on the other hand, prevailed before the
bisection, why was the bisection made, leaving the intermarrying
and exogamous pairs of totems precisely in their old position ?
What valuable novelty was introduced by the bisection?

Suppose, on the other hand, that the phratries (as in my
hypotliesis) result from an alliance, with connubium^ between animal-
named groups previously independent and probably hostile; each
local group containing members by female descent of several other
animal-named groups. For what conceivable purpose, in that case,
were the scattered members of so many groups obliged to be
exogamous and intermarrying in pairs of totem kins, and never
out of the opposed unit of the pair ?

I can, it is true, conceive a state of affairs which might
produce, first, the one totem to one totem rule ; next, the phratry
divisions, and the partition of the totems between the phratries.
But though the state of affairs whence these results might naturally
flow is conceivable, and possible, it seems far from probable.
However, I give it,

1< Suppose local groups of animal names, exogamous and
hostile.

2. Suppose that captive women do not retain and bequeath
to their children, their original group names. The Emu group, Bee
group, &c., consists wholly of Emus, Bees, and so on in each case,

3. Becoming necessarily consanguineous, each pair of groups
become friendly, and establish peaceful connuUum, Snake with
Swan, and so on.

4. All one set of paired groups become allies, taking a name,
say Eagle Hawk, for their little confederacy, but still, like many
extant tribes, observing the one totem to one totem rule in
marriage. The corresponding set of groups (B) passes through
the same processes, and finally strikes up friendship with Group A.
Call the second group Crow.

AUSTRALIAN PROBLEMS

207

5. We now have a tribe, consisting of Phratry A (Eagle Hawk)
and Phratry B (Crow). They retain these names; and thus are
a MuktvararKUpara tribe. But the pairs^ say Swan in Eagle Hawk,
Snake in Crow, still (as among the Northern Urabunna) intermarry
only with each other. The arrangement is stupid, but not more so
than the Northern Urabunna custom.

We have then one totem to one totem marriage. A difficulty
is that I cannot see why, on this supposition, the names of the
two groups, say Dog and Duck, should descend in the female line.
I have satisfied myself, at least, with an explanation of that pecu-
harity, in the system set forth in The Secret of the Toteni, but I can
imagine no cause of it on this theory. Or is this the reason?
The two, say intermarryhig groups, Dog and Duck, were in
separate localities. The object is to interfuse them in amity. If
male Dog marries female Duck, and the children taking the
paternal totem name are Dogs, and if when male Duck marries
female Dog the children are Ducks, then Dog group and Duck
group are not interfused ; Dogland is Dogland, Duckland is Duck-
land for ever; the groups remain separate entities. Tliis is
avoided by the reckoning of descent in the female line.

This new suggestion, like my suggestion in The Secret of the
Totem, gives a natural human cause for the phratries, and for the
presence of totem kins within the pln-atries, while it accounts for
the alleged one totem to one totem, and one set of totems to
another set of totems, surviving from an older age into that of the
phratries of to-day.

I have generously presented a guess to explain the origin
of one totem to one totem marriages, on the supposition that the
institution really exists. Perhaps it is not invidious to say that
Mr. Spencer's assertion of the emphasis with which the northern
Urabunna attest its existence is more satisfactory evidence than the
testimony of Mr. Boultbee to the same rule in the Itchumundi,
Karamundi, and Barkinji (?) nations, owing to the great experience
of Mr, Spencer. In matters so difficult, however, and in languages
where the words for totem, and 'class* (phratry) are apt to be
the same, it is not impossible that even Messrs. Spencer and
Gillen may have been misunderstood, and therefore misinformed,
by their native friends.

The idea that there exists an institution limiting marriage of
members of one totem kin only to members of one totem kin only

208

AUSTRALIAN PROBLEMS

in the opposite phratry, or exogaoious half of the tribe ; and the idea
that, in other tribes^ three or four totem kins in one phratry are
limited to marriage with three or four totem kins only in the
opposite phratry, is by no means new. In Messi^s. Fison s and
Howitt*s Kamilaroi and Kumai (1880) Mr, Fison quotes Mr. Lockhart,
who was intimately familiar with the Mukwara-Kilpara tribes of
the Darling River more than fifty yeai's ago. Mr. Lockhart wrote
(no date is given), ' The females of the Wild Duck (totem), we shall
say, are aU (by phratry) Kilparas (Crow), and they take Mookwara
men of the Emu/

If Mr, Lockhart means that men of the Emu totem, of Mukwara
phratry, married only women of the Wild Duck totem in Kilpara
phratry, we have here one totem to one totem marriage. But he
goes on, * The Kilpara Wild Duck boys look out for, say, Mookwara
Emu girls.' The word * say ' throws doubt on his intention to assert
that the Kilpara Wild Duck boys could marry only Muquara Emu
girls, and no other Mukwara girls (Kam. and Kur.^ p. 43).

Mr. Fison does not seem to have pressed his inquiries to the
point of asking whether Mr. Lockhart really asserted the existence
of one totem to one totem marriage. He did learn, from Mr.
Stewart, that any man of any totem of Kumite (Cockatoo) phratry
in the Mount Gambier tribe, could marry a woman of any totem
in the Kroki phi'atry of the same tribe, as among many tribes.

Mr. Fison was strongly of opinion that, in several tribes,
marriage was limited to men of X totems of A phratry with
women of X totems in B phratry. Mr. Daniel M'Lennan, in his
brother's Studies in Ancient History (p. 596, Second Series), spoke of
Mr. Fison's belief * that where a division ' (phratry) * includes several
totems, and there is no marriage within the division * (phratry) ' the
totem puts a further restriction on marriage ' — as a mere dream*
Such evidence as we receive from Mr. Homtt and Messrs. Spencer
and Gillen confirms Mr, Fison*8 belief (if he entertained it) that one
totem to one totem marriage does exist in many tribes. Mr. Fison
even held that sets of totems intermarriageable only with each
other, in phratries Kilpara and Mukwara, are ' analogous ' to the
four sub-classes of the Kamilaroi, and to the four subclasses
of Queensland tribes with the phratries Yungaru and Wutaru.
(A' and K,, pp. 41-2.) Mr. Fiaon, I tliink, was arguing in the right
direction. But he held that the phratries came first of all, and were
later * subdivided * into sets of totems stiU further limiting marriage.

AUSTRALIAN PROBLEMS

209

On this head, as I cannot make out what Mr. Fiaon conceived the
origin of totem kins to be (apparently he thought them deliberately
introduced ' classes ' intended to Umit marriage, though why they
bear animal names he saith not), it is not necessary to say more.

Let us next try how one totem to one totem marriage can be
explained on the theory of bisection,

A previously * undivided conmiunal horde ' is legislatively
bisected into two exogamous intermarrying divisions, say Crow
and Eagle Hawk.

Why is this done ?

Is it (1) to satisfy * tribal morality *, and, if so, how has the
* tribe ' become moral ?

Is it (2) to limit by 50 per cent the chances of murderous
rivalries in love?

Is it (3) to prevent some supposed ill consequence to inter*
marrjdng uterine brothers and sisters— and also to * tribal ' (or rather
phratria^) brothers and sisters ?

Is it (4) from a felt need of organization of one kind or another,
at random ?

Are there already totems in the undivided community, and, if
soj how can the horde be called undivided ? We must mean matri-
monially undivided.

Are there no totems, and, if so, whence and wherefore did they
come ? Prom magical societies ? I can point out the difficulties of
this idea

If there are totems, why are they so arranged that the same totem
is never on both sides of the division (save among the Arunta) ? '

Why have phratries of translatable names got animal names ?

Were the opposed totems, from the first, intermarriageable only
in pairs, and, if so, why ?

Were they originally all intermarriageable with all totems on
the opposite side ?

If so, why did certain vei-y primitive tribes later compel one
totem of one side to marry only one totem on the other side ? Mr.
Howitt writes ' (p. 189) : * Tlie restriction in mamage to one or

^ I must g'ive the whole of Mr, Howitt's sentence, which I may mis-
apprehend. ' The restriction in marriage to one or more totems is certainly later
in origin than the Bieri rule, and, as will he seen hy my further statements^
in many tribes with two-class divisions/ I miderstand Mr. Howitt to mean
' later than the Dieri rule^ and than the rule which prevails in many other tribes
with onltf two^lass diviaions*.

210

AUSTRALIAN PROBLEMS

more totems is clearly later in origin than the Dieri rule/ * Sub-
classes ' are universally supposed to be later, but what proves that
the one totem to one totem rule is later ?

We seem lost in a wilderness of diflSculties I

n* Animism and Nescience of Procreation,

When heaven, to punish the sins of the learned, permitted
Messrs- Spencer and Gillen to discover and describe tlie institutions
of the Arunta nation, it was found that these tribes practised an
unheard-of kind of totemism. The totem was not hereditary, the
totem-set of people in each case was not exogamous ; a man who
waB a Dog might marry a woman who was a Dog, and their children
might be Eat, Cat, and Frog.

Instantly some of the learned (A) averred that this unheard-of
form of totemism was tlie oldest extant end the nearest to the
primitive model ; wliile others (B) declared that the Arunta totemism
was a decadent * sport', and showed how it arose, or might have
arisen, out of exogamous totemism and hereditary totems. I was
of the second party, the B division, from the first.

The A division, who regarded Arunta totemism as the earliest,
naturally tried to show that, in other matters, the Arunta nation was
the most primitive. The Arunta wore no clothes, and they were
ignorant of the fact that sexual connexions are the cause of concep-
tion and birth — what coiM he more primitive ? They also practised
'co-operative totemic magic ; and co-operation, duly organized, may be
more primitive than individual effort ; the division of labour being
also primitive.

To this the opposite faction (B) replied that the Ai*unta (1)
exhibited confessedly the most complex, and, as had hitherto been
agreed, the latest form of matrimonial rules, the * eight-class
system \ Next (2), they reckon descent and transmit hereditable
property in the male line, and hitherto we had unanimously
supposed reckoning in the female line to be the earlier. Next
(3), they had lost the names of their primaiy exogamous divisions
(phratries), and, hitherto, these names had been looked on as very
early. Next (4), they practised the bloody rites of initiation which
Mr. Spencer tliioks posterior in evolution to the south-eastern
dentistry. (5) The Arunta have no * All Father ', and while the A

AUSTKALTAN PROBLEMS

211

disputants thought this a proof of primitiveness^ the B party held
that the animistic philosophy of the Arunta had left no logical ramn
d'etre for a creative ' All Father \ (6) The B faction held that
co-operation and division of laboui*, each totem-set doing magic for
its own totem, were not primitive, but much the reverse.

The A party admitted the social advance of the Aiiinta to the
eight^class system. Advance, however, is not uniform ; a tribe
might reach the eightrclass system, but be primitive in other
respects. As to Arunta male descent (hitherto looked on as a proof
of advance), the A party suggested that one tribe might begin with
male descent, and another with female descent, though we have
irrefiitable proof that, in other northern tribes of the eight-class
system, female descent has left indelible traces, and no proof that
male descent has ever become female descent.

The Arunta philosophy of reincarnated spirits, entertained, with
modifications, by tribes of female descent near Lake Eyre, and by the
northern tribes, with male descent, is entirely animistic. Among
the Arunta, at the beginning of things, rudimentary animated bulks
of lacustrine environment were converted by two beings named
* Self-€!xisting ' or *Made out of Nothing' (Ungambikula) into
animal forms of known species. One of these beings might be
styled either * a man-kangaroo * or * a kangaroo-man \ They went
about playing their pranks and founding rites and institutions,
carrying decorated stone plaques^ called churinga, still used by the
Arunta. Their bodies died, but their immortal part haunted the
stone churifiga. Tliese immortal spirits, the Arunta say, cause
conception by entering into women who pass the places where the
churinga were deposited. Thus every Arunta has been, in the spirit,
from the beginning, and will endlessly be reincarnated. Conse-
quently sexual connexion does not cause conception and birth. ^
How could it? A baby is to the Anmta only a being who has been
from the beginning — now in the flesh, and now out of it— and who
will so continue to be. Such a spiritual entity cannot conceivably
owe his existence to gross material amours. The thing, to an
Arunta philosopher, is unthinkable. For this pliilosophic reason,
says party B, the Arunta ignore procreation, A man cannot beget
an everlasting spirit * No,' says party A ; 'the Arunta are too
primitive to understand physical processes wliich are sufficiently
understood by other savages/

* Spencer and Gillen, (kniml Tribes, p. 265,
P 2

212

AUSTRALIAN PROBLEMS

For example, certain tribes of southeastern Australia,
including some who reckon descent in the female line^ hold that
* children originate solely from the male parent, and only owe their
infantine nurtui-e to their mother '. * A woman is only a nurse who
takes care of a man's cliildren for him/ * But matrimonial life,
among the Arunta, is supposed, at most, only to prepare a woman
for the entrance of a spirit wliich has existed from the beginning,
the Alcheringa, Therefore the Arunta are in pristine ignorance of
physiology,

I have argued, often and in many places, that the Arunta
nescience of the part of the male in procreation need not be a proof
of absolute ignorant ' primitiveness \ but merely the logical result of
their animistic philosophy. Their psychology has clouded their
physiology. Every one of them, according to their elaborate
philosophy (which surely no mortal can think * primitive '), has
existed since the beginning and can never cease to exist.

No efforts of men and women can produce a spirit which, they
say, is pre-existent and of endless existence. The logical black
fellow, granting his premises, can come to no other conclusion than
that human beings — ^incarnate spirits— do not beget pre-existing
spirits. Their specnlations deal with the spirit, fornm fomuifis^
neglecting to account for the body of flesh. There is nothing
' primitive ' in all this ; there is only logic working on the basis of
animism, or so it seems to me.

As far as I am aware, nobody except M. van Gennep, who
beUeves in Arunta ^ primitiveness ' has tried to meet my argument,
or even made it the subject of an allusion. =^ But I have seen many
grateful references to Dr. Roth's discovery of denial of human pro-
creation by other tribes, a discovery set forth in liis Bulletins on
North Queensland Ethnography (No. B, 1903). In most references to
Dr. Roth which I have seen, the details of his discoveries were
not fiiUy discussed. 1 therefore discuss them ; they show tliat an
animistic philosophy, differing in many poiats from that of the
Arunta, colours and even causes the Northern Queensland denial of
procreation. When North Queensland peoples say that the lower
animals have no spirits or souls, and that they may be and are the

» Howitt, J. A. /., 1882, p. 502 ; N. T, S. E. A., pp. 283-4.

' M. van Genuep, indeed, urges that the spirits of the Alcheringa folk have
not existed from the beginning* They are as old as the beginning, for the
Alcheringa is the beginning. (M^tfies el Legendes d'Austmlie, p. Ixv, note 8.)

AUSTRALIAN PROBLEMS

213

result of procreation ; whereas mankind, having spirits, are not and
cannot be procreated, but are made or created, then we have to
confess that, in the case of mankind, the North Queensland
l^ychology hixs clouded the Queensland physiology. The North
Queensland tribes know the method of the procreation of the lower
animals. What they deny is that physical procreative processes can
produce man, who has a soul, who is a living spirit, I have been
unaware that the Queensland blacks draw this essential and illumi-
nating distinction between man and beast, because, till lately,
I had never been able to procure Dr. Roth's Bulletin No. 6.

Dr. Roth says, * Animals and plants are not regarded as having
any *' Koi " — spirit or soul.' ^ * Although sexual connection as a cause
of conception is not recognized among the Tully River blacks so far
as they themselves are concerned, it is admitted as true for all
animals ; indeed this idea confirms them in their belief of superiority
over the brute creation.' Connexion can make a brute ; * to make a
mufis beyond its might,' as Burns says, for man is a living spirit/^

These passages prove, I hold, beyond possibility of doubt, that
the animistic or spiritual philosophy of these blacks, and nothing efee,
causes them to deny that sexual connexion is the agency in the
making of man. They have to invent other ways.

I have always conjectured that * These things are not otherwise,
but thus', that psychology has darkened physiology among these
logical thinkers, and now my conjectui-e is established, as far as Dr.
Roth's tribes are concerned. They understand the physical pro-
cesses of procreation among all the other animals, but these animals
have no soul or spirit, have nothing immaterial. Man, on the other
hand, has an essential, immaterial part, his Koi, or soul. The
physical and material processes which account for the reproduction
of soulless brutes cannot account for the births of spiritual men.
They must * come otherwise '. Manifestly the North Queensland
blacks have come across an old enigma of speculation, which they
solve in their own way. The enigma is, How could beings clearly
material acquire an immaterial indwelling spirit ?

Even Dr. Roth, who gives us the facts, does not appear always
to understand their bearing on the denial of procreation among
men and women. He writes (Bulletin No. 6, pp. 23, ^ 83) : * When
it iff remembered that as a rule in all these Northern tribes a little

' BuUctln No. 5, pp. 17, § 64.
■ mtUetin No* 5, pp. 22, $ 81.

214

AUSTRALIAN PROBLEMS

girl may be given to and will live with her spouse as wife, long
before she reaches the age of puberty^-the relationship of which
to fecundity is not recognized — the idea of conception not being
necessarily due to sexual connection becomes partly intelligible,'
But nothing has been said about human conception not being
necessarily due to sexual connexion* It has been said that human
conception cannot be due to connexion at all ; for the reason that
material processes cannot produce a koi or soul, or spirit. This
opinion is not * partly intelligible', but entirely intelligible, and,
given the black's premisses, is his only logical conclusion.

How a human child comes the blacks tell us, and Dr, Roth
gives us their theory. Just as the Eternal, in Genesis, formed
a figure of clay, and breathed into it a Uving spirit, so, according
to the blacks, before each new human birth, an invisible being
makes a doll of mud, informs it with a Uving spirit, and conveys
it into the womb of its mother.

This process — men are not created, nor begotten, but made—
and not material procreation, accounts among these blacks for
human births. Meanwhile the soulless animals reproduce each
other in the usual way. This conclusion, or something like it, is
forced upon the black thinker by his belief that man is a body
enshrining a living spirit, wliile a brute is a body enshrining nothing
of the sort ; the beasts perish.

This pliilosophy is the reverse of * primitive ' ; it does not indi-
cate pristine ignorance, but the logical invention of the spirituahst
philosopher. It took him long, doubtless, to evolve the idea of
spirit from his experiences of dreams, trances, coincidental death-
wraiths, hallucinatory phantasms of the dead, crystal-gazing, vue a
distance^ or clairvoyance, and hypnotism. Man was not ^ primitive '
when he had amassed and speculated on all these experiences, and
had recognized himself as a spiritual being, encased in clay* He
was not primitive when he patiently and logically worked out his
complex animistic philosophies, varying in different tribes. Some
south-eastern tribes have not worked out their psychology to its
necessary conclusions ; they have discovered the physical causes
of procreation, and do not trouble to inquire, * Whence and
how comes the informing spirit?' The northern nations, on the
other hand, have resolutely pushed their animism to its necessary
conclusion, and deny that material processes produce the spirit
of man.

ADSTRALUN PEOBLEMS

216

Let us consider in detail the psychology of our dark fellow
subjects. On the TiiUy River the spirit {Kot) is associated with
shadow and breath ; is intangible (' a spirit hath not flesh and
bones % practises rapping (like other spooks) after leaving the body,
haunts its old home and friends, finally its address is ' The Bush \
It is everlasting (' so far as the blacks have any conception of the
term '), does not feed, hiis no gifts or sacrifices of food ; and is rather
dangerous to persons whom it encounters. Phantasms of the dead
may be seen by individuals, or in a collective hallucination expe-
rienced by several persons at once. But * animals and plants have
no Koi\ {Query— ^ KoMn, a glorified and deified black fellow',
cf. Howitt, pp. 497-99.) On the Blomfield Eiver, and at Cape
Bedford, dogs have * thinking powers \ and have wau^tvUj which is
equivalent to Koi; the term is apparently an onomatopoeic word
for ' breath \

On the Pennefather River, Ngai and Cho-i {Kot) are connected
with the heart and the * after-birth ', not with the bre^ith. {Gko-i
seems to be Koi^ I think.) No one has a iigai till the death of
his father by blood, when the paternal ngai passes into all his
children. When none of the kin are left alive, the ngai * finishes
altogether '. But every one has also a cho-i^ which a m3rtlucal beings
Anje-aj the baby-maker, puts into a doU of mud, before inserting
the doll, spirit and all, in the maternal body. The cho-i^ like the
Arunta akheringa spirit, is constantly reincarnated. Part of it
leaves the body at death, and becomes a wandering phantasm, but
another part remains in the * after^birth *. The lifter-birth is buried
by the grandmother in a marked place, whence Anje^ takes the
cho'i^ and he keeps it in a lagoon, a rock crevice, or a tree, tiU
he needs it. When baby-making, he takes the spirit or cho-i from
its hiding-place, inserts it in his mud doll, and puts the mud doll,
ck(hi and all, in the maternal body* The child's hunting-ground
is the region where Anje-a kept it till he thus used it. ' Animals
and plants have neither n^fai nor cko-i/ and procreate in the re-
cognized way. The names for cho-i^ and for the supernormal
baby-maker, vary in different tribes. The cho-i is thus permanently
existent, and perpetually reincarnated, hke the akheringa spirits
among the Arunta

This behef is much akin to that of the Eualilayi, where Moon
and Crow are the baby-makers. But the Eualilayi do not admit
reincarnation, except in the case of the spirits of children who

216

AUSTRALIAN PROBLEMS

die uninitiated. The spirits of the adult dead go to their own
place, good or evil, or are reincarnated in birds. The Euahlayi
recognize the advantages of the procreative assistance of the male.
* To live, a child mmt have an earthly father ; that it has not, is
known by its being born with teeth.' ^

The Queensland philosophy has this advantage over the Arunta
theory, that it accounts for the material body in which the re*
incarnated spirit appears. The body is made of mud, or of pandanus
roots, by the baby-maker, Anje-a^ or Nguta Ngtita^ alias Taipan^ or
Kunya^ all of them beings who haunt wood and water. On the
Tully River babies also come by suggestion, ' some man may have
told the woman to be in an interesting situation,' or in consequence
of a dream » or she has caught a bull frog, or sat over a fire on
which she has raasted a bream given to her by the child's * pro-
spective father". Babies are made beyond the sunset, and enter
the mother as plovers or snakes, resuming their human form when
settled.

It appears that there is considerable variety in the hypotheses
of these blacks as to birth. They have not the serene orthodoxy
of the Arunta. However, their reason for denying procreation
in the human race, while admitting its existence among the lower
animals, has been made conspicuous. These people do not err from
ignorance, but in obedience to their philosophy of animism.

If the Arunta suppose that even among the brutes conception
has not its natural cause, the reason is also philosopliic. All spirits
were during the Alcheringa encased in animal forms, man-kangaroo
or kangaroo-man. A kangaroo totem spirit therefore may as readily
enter a female kangaroo as a woman of the kangaroo totem. But
I am not positively certain that a female kangaroo is believed never
to become pregnant except through the invasion of a spirit from
a kangaroo OhmnikiUa^ or mortuary local totem centre.

My conclusion is that among tribes who believe in peipetual
reincarnation of each practically everlasting spirit, the denial of
procreation is not a * proof of pristine ignorance *, but a philosophic
inference from philosopliic premisses. If I am right * conceptional
totemism' among the Arunta cannot be pristine. It is a theory
logically drawn from the pliilosophic conception that each unceasing
spirit was originally a totemic spirit, and continues to be the same
totemic spirit through all its reincarnations. Where savage thinkers
* Mrs. Langloh Parker, Tim Euahlayi Ttibe, pp. 50-L

AUSTRALIAN PROBLEMS

217

have not worked out this or some similar philosophical theory, there
is no * conceptional totemism ', as far as I am aware. If we would
argue that conceptional totemism is the earliest, and has been the
universal, form of the institution, we must assume that raankindj
everywhere, once held that every human being bom is bom of
the spirit ; and that mankind, in all known places except some
regions in Australia, abandoned that opinion. Proof of this
hypothesis there can be none, and such a universal uniformity
of adhesion to one of the endless forms and phases of animistic
philosophy is contrary to all that we know of philosophers, early
or late. They are never all agreed in one system. All known
savages have reached the belief in spirits, but, as far as we are aware,
not all savages have applied the spiritual theory in their myths
of the origin of totemism* That application, as far as we know,
has only been made in Australia, by many tribes, and various sets
of tribes differ much in the form of their theory. Some, having
always made totems hereditary and exogamous, retain that system,
and explain it by various forms of the animistic behef. Others, having
drifted, for obvious reasons, into non-hereditary, and non-exogamous
totemism, plus the stone churinga creed, have accounted for all
that by another twist of speculation ; these are the Kaitish and
Arunta.

Their myth was never universal ; it is limited to the region in
which stone ehttringa are common implements.

So far I had written, with a good deal of confidence, when my
attention was drawn, by Mr. N. W. Thomas, to the following
passage from a letter by the Rev, Mr, Strehlow, who has long Uved
as a missionary among the Dieri and the southern Arunta, and is
able to speak and write the languages of their tribes. He says
{Globu^^ vol. xci, p, 289) : ^ If a woman perceives the first signs
of pregnancy immediately after seeing a kangaroo which disappears
before her eyes, she is certain to become the mother of a kangaroo
child. If the signs are perceived after a hearty meal of some fruit,
she has conceived a child which has that fruit as totem.' TMs
statement precisely confirms the theory of Mr. J. O, Frazer stated
by him in the Fortnightly ItevieWj September, 1905, and is, perhaps,
the most surprising anticipation of facts later discovered by re-
seai*ch that has been made since the theoretical finding of Neptune.
But there are some points as to which I am uncertain. Is a sjnrit
supposed to emanate from the kangaroo or the hearty meal of fruit ?

218

AUSTRALIAN PROBLEMS

As the Arunta nation notoriously get their totem otherwise, namely
from totemic spirits haunting ohianikilh^ is this myth a sporadic
variant? No search, one may presume, is made for the chumiga
nanja of the child when born under this variant myth. Do the
people who entertain the myth vary, in any other ways, fi^ora
Arunta orthodoxy? Mr, Frajzer added, among other adminicles of
evidence in favour of his theory, that it explained why the great
mfyority of Central Australian totems are edible objects. But that
is not strange, for the Central Austrahan finds all plants and animals
edible, except a very few insects (Northern Tribes^ pp. 768-73). They
can easily eat Karti, a full-grown man, or Thaballa, a laughing boy,
totems of the Warramunga and Tjingilli. They cannot eat the
Wollungua, a totem, nor even see him, as he does not exist. So
how did he thrive to totem's estate, unless, perhaps, a woman
thought she saw Mm ?

Tlie argument is not aided by the intermingling in the same
community of men and women of many different totem stocks, for
that is caused by exogamy with female descent, and again is caused
by the Oknanikilla system, among the Arunta. The strength of the
argument for conceptional totemism is the discovery by Herr Strelilow
of its actual existence. But is the mylih animistic? If so, is it
probable that all totemic mankind have held by this phase of
animistic philosophy? Here we are once more in the region of
probabilities. It remains odd that no woman or girl totem, and
only two human totems, are known, though mothers are just as
likely to have seen females of their species at the critical moment,
as a kangaroo, or a dish of fruit.

The theory of conceptional totemism rests, as Mr. Strehlow's
evidence proves, on a vera cama. There is no absolute reason, prima
facie, why conceptional totemism should not have been universally
the origin of the belief that children were actually identical with the
various objects in nature whose names and natures they acquired at
birth. But, if we start from the hypothesis that these processes
have been primary and universal, we encounter difficulties in tracing
the further evolution of totemic institutions which cannot be
discussed here, but which, to myself, seem insuperable, at least on
the theory of the legislative bisection of a community containing
animal-named sets of kindred, or magical societies, not previously
exogamous.

IS TABOO A NEGATIVE MAGIC?

By R R MARETT
Fellow of Exetee College

It is always easier to criticize than to construct. Many affirma-
tive instances usually go to the founding of an induction, whereas
a single contradictory case suffices to upset it. Meanwhile, in
anthropology, it will not do to press a generalization overmuch, for
at least two reasons. The first of these reasons is the fundamental
one that human history cannot be sliown, or at any rate has not
hitherto been shown ^ to he subject to hard-and-fast laws. Hence
we must cut our coat according to our cloth, and be fully content if
our analysis of the ways and doings of man discloses tendencies of a
well-marked kind. The second reason is that, in tlie present state of
the science, field-work, rightly enough, predominates over study-
work* Whilst the weather lasts and the crop is still left standing,
garnering rather than threshing must remain the order of the day.
Working hypotheses, therefore, the invention of theorists who are
masters of their subject, are not so plentiful that we can aflFord
to discard them at the first hint of an exception. If, then, some one
comes forward to attack a leading view, it is not enough to arm
himself with a few negative instances* It is likewise incumbent on
the critic to provide another view that can serve as a substitute. In
the present case I have sought to do this after a fashion, though I
am painfully aware that, in defining taboo by means of nmna, I am
laying myself open to a charge of explaining obscurum per obscurim.
I can only reply prophetically that the last word about mana has not
yet been said ; that it represents a genuine idea of the primitive
mind, an idea no less genuine and no less widely distributed than
the idea of taboo, as several writers have recently suggested, and as
further investigation will, I beUeve, abundantly confirm, I would
also rejoin that if the accusation of obscurum per obscurius hardly
applies directly to the theory I am criticizing— since to identify
* magic * with the sympathetic principle yields a perfectly definite
sense — yet the natural associations of the word are so much at
variance witli this abstract use of the name of a social institution

220

IS TABOO A NEGATIVE MAGIC?

that the expression * negative magic * is more likely to cause confusioa
than to clear it up*

So far back as when Dr. Tylor published his epoch-making
MesearcJies into the Early History of Mankind we find the suggestion
put forward of a certain community of principle between taboo and
that ' confusion of objective with subjective connexion * which ' may
be applied to explain one branch after another of the arts of the
sorcerer and diviner, till it almost seems as though we were coming
near the end of his list, and might set down practices not based on
this mental process as exceptions to a general rule/ ^ * Many of the
food prejudices of savage races,' continues Dr, Tylor, * depend on
the belief wluch belongs to this class of superstitions, that the
quaUties of the eaten pass into the eaten Thus, among the Dayaks,
young men sometimes abstain from the flesh of deer, lest it shoidd
make them timid, and before a pig-hunt they avoid oil, lest the
game should slip through their fingers, and in the same way the
fliesh of sloW'going and cowardly animals is not to be eaten by
the warriors of South America ; but they love the meat of tigers,
stags, and boars, for courage and speed/ *

Recently '^ Dr. Frazer has univeraalized Dr, Tyler's partial
correlation, and has pronounced ^ the whole doctrine of taboo ' to be
a negative magic, understanding by magic a misapplication of the
association of ideas by similarity and contiguity. A very similar
definition had already been proposed by MM, Hubert and Mauss.*
Tliey limit the identification, however, to what they name ' sympa-
thetic taboo ', implying that taboo includes other varieties as well.
Again, although here they seem to make the sympathetic principle
the differentia of magic, the final gist of their admirable essay is
rather to find this in the anti-social character ascribed to the
magician's art.

Now, according to the foregoing view, taboo is a ceremonial
abstinence based on the fear of definite consequences. Just as
sjrmpathetic magic says, * Aa I do this, so may that which this
symbolizes follow,' taboo says, ' I must not do this, lest there follow
tliat which is the counterpart of this/

In violent contrast we have the view of Dr. Jevons, wliich, at

' Op, cU., Brd edit., 129. * ib. 131.

^ Lectures on the Early History qf tlie Kingship^ 62.

* L'Ann^ Sociologiqttt, viL 56. It is to he noted that Dr. Frazer arrived at
his conclusion by independent means ; cf, Man, 1906, 87,

IS TABOO A NEGATIVE MAGIC?

221

first sight at any rate, seems to declare all consideration of conse-
quences to be foreign to the taboo attitude. He bases his theory
of taboo on an alleged * fact that among savages universally there are
some things which categorically and unconditionally must not be
done % insisting * that this feeling is a ** primitive *' sentiment *. ^ Now
it is not easy to discover what is here meant, so great is the
departure from the recognized terminology of philosophy. ' Cate-
gorically" and * unconditionally ' are expressions that smack of
Kantian * rigorism ' ; but Kant's famous analysis of duty as a
categorical and unconditional imperative makes obMgation directly
antagonistic to sentiment of all kinds. A sentiment as such has
a history and assignable developmeiit* The Kantian law of duty,
apriori, objective, absolute, has none whatever. Is Dr, Jevons, then,
speaking here strictly according to philosophic tradition ? Or would
he recognize a growth of moral principle, say, on some such lines as
those which Dr. Westermarck or Mr. Hobhouse has recently laid
down ? If he were of the former persuasion, then he would be
irrelevantly interpolating a non-genetic view of moraUty that for
purposes of psychological and sociological explanation could have no
value or significance at all. But if he is of the other and less uncom-
promising faith — which appears more probable, seeing that his book
is professedly dealing with religion from the historical standpoint —
then ' categorical * and * unconditional ', in their application to a mere
sentiment, are to be given an elastic sense. No more is meant, we
must in that case suppose, than that the taboo feeling of * Do not
meddle ' involves no very expUcit condition, no very clear or specific
idea of unpleasant consequences to be avoided, but as it were
threatens by aposiopesis — * Do not meddle, or, if you do, . . • !' If
this is as much as Dr. Jevons intends— and it seems at any rate to
be all that is meant by MM, Hubert and Mauss when they speak in
very similar terms of the absolute, necessary, and a priori character
of the * magical judgement ' ^ — then I think this view has very much
to be said for it

My own contention is that, whilst there is always a sanction
at the back of taboo in the shape of some suggestion of mystic
punishment following on a breach of the customary rule, yet
the nature of the visitation in store for the offender is never
a measurable quantity. Even when the penalty is apparently

* An Inirodtiction to iJm Eistortf of Meltgion, 85.

* Op. eiL, vii. 125.

222

IS TABOO A NEGATIVE MAGIC?

determinate and specific — which, however, is by no means always
the case, as I shall endeavour to show later— an infinite plus of
awfulness will, I believe, be fonnd, on closer examination, to attach
to it Taboo, on my view, belongs, and belongs wholly, to the
sphere of the magico-religious. Within that sphere, I venture to
assert, man always feels himself to be in contact with powers whose
modes of action transcend the ordinary and calculable. Though he
does not on that account desist from attempting to exploit these
powers, yet it is with no assurance of limited liability that he enters
on the undertaking. In short, dealings with whatever has mystic
power are conducted at an indefinite risk ; and taboo but embodies
the resolution to take no unnecessary risks of this indefinite kind.
This contention I shall now try to make good.

First, to attack the theory that taboo is negative magic (in
Dr. Frazer's sense of the term * magic ') on the side on which that
theory is strongest, namely where sympathy is most in evidence.
I do not for one moment deny that in some taboos a sympathetic
element is present and even prominent. Indeed, I see no harm in
speaking, with MM. Hubert and Mauss, of sympathetic taboo, where
* sympathetic ' stands for the differentia or leading character of
a variety, and the genus * taboo ' is taken as already explained in
independent terms. The presence of the sympathetic principle is,
to my mind, amply and crucially proved in the case of those food
restrictions mentioned in the passage quoted from Dn Tylor, the
prohibition to eat deer lest one become timid, and so on. Another
teUing set of examples is provided by those remarkable taboos on
the use of knots which, as Dr. Frazer has abundantly shown, are
wont to be observed at critical seasons such as those of child-birth,
marriage, and death J But even here, I suggest, the consequences
tend to remain indefinite and vague, and that for more than one
kind of reason.

We can distinguish a sociological reason and a hierological or
religious reason, though for the purposes of the historical study of
rehgion, from the standpoint of which taboo is usually considered,
the first may be treated as subordinate to the second.

To begin with, these, no less than any other taboos, are
customary observances, a portion of the unwritten law of society.
To this fact, then, must be ascribed part at least of the force that
renders them effective. There are always penalties of a distinctively

' Tiie Golden Bough % i. 392 sqq*

IS TABOO A NEGATIVE MAGIC?

228

social kind to be feared by the taboo-breaker. In extreme cases
death will be inflicted ; in all cases there will be more or less of
what the Australian natives call * growling V and to bear up against
public opinion is notoriously the last thing of which the savage
is capable. Moreover, this social sanction is at the same time
a religious sanction. To speak the language of a more advanced
culture, State and Church being indivisibly one, to be outlaw
is ipso facto to be excommunicate. Given the notion of mystic
danger — of which more anon — social disapproval of aU kinds
will tend to borrow the tone and colour of religious aversion, the
feeling that the oflFender is a source of spiritual peril to the com-
munity ; whilst the sanctioning power remains social in the sense
that society takes forcible means to remove the curse from its midst.
It may be argued that these social consequences of taboo-
breaking are secondary, and thus scarcely bear on the question of
the intrinsic nature of taboo. Such an objection, however, will not
be admitted by any one who has reflected at all deeply on the
psychology of religion. On the broadest of theoretical grounds
religion must be pronounced a product of the corporate life—
a phenomenon of intercourse. Confirmation a posteriori is obtained
by the examination of any particular taboos of w^hieh we have
detailed information. Take, for example, the elaborate Ust of food-
restrictions imposed amongst the Arunta on the ulpmerku or boy
who has not yet been circumcised.* The sympathetic principle is
probably not absent, though its action happens here not to be easily
recognizable. When we learn, however, that eating parrots or
cockatoos will produce a hollow on the top of the head and a hole
in the chin, we may suspect that the penalty consists in becoming
like a parrot or cockatoo. On the other hand^ the same penalty, for
instance premature old age, follows on so many different kinds of
transgression that it looks here as if a tendency to dispense with
particular connexions and generalize the effects of mystic wrong-
doing were at work. Meanwhile^ in regard to all these taboos alike
our authorities assure us that the underlying idea throughout is that
of reserving the best kinds of food for the use of the elder men, and
of thereby discipHning the novice and teaching him to ' know his

' Cf. B, Spencer and R Oillen, Tfte Natk^ Tribes of Central Amtralia, 196.

^ Spencer and Qillen, op, ciL^ 470 aqq. Here, by the way, in the systematic
aasignmeni of penalties to offences we aeem to have a cniciAl disproof of the pure
* unconditional ity ' of taboo.

IS TABOO A NEGATIVE MAGIC?

place*. Here is a social reason with a vengeance. Even if sonae^
suspect that our authorities over-estimate the influence of conscious
design upon tribal custom, they will hardly go the length of assert-
ing that sympathy pure and simple has automatically generated
a code so favourable to the elderly gourmet. A number of succulent
meats to be reserved on the one hand, a number of diseases and
malformations held in dread by the tribe on the other, and possibly
a few sympathetic connexions established by tradition between
certain foods and certain diseases to serve them as a pattern — with
this as their pre-existing material the Australian greybeards, from
all we know about them, would be quite capable of constructing
a taboo-system, the efficient cause of which is not so much mystic
fear as statecraft Even if the principle of sympathy lurk in the
background, we may be sure that the elders are not applying it very
consciously or very strictly ; and again we may be sure that society
in imposing its law on the ulpmerka is at much greater pains to
make it clear that he must not eat such and such than why he
must not^if only because there are so many excellent reasons of
a social kind why the young should not ask questions, but simply
do as they are bidden.

But there is, I believe, another and a deeper reafion why
sympatliy pure and simple cannot account for taboo. Taboo, I take
it, is always something of a mystic afltair. But I cannot see why
there should be anything mystic about sympathy understood, as
Dr. Frazer understands it, simply as a misapplication of the laws of
the association of ideas. After all, the association of ideas is at the
back of all our thinking (though by itself it will not account for any
of our thinking) ; and thinking as such does not fall within the
sphere of the mystic* Or does the mystery follow from the fact that
it is a * misapplication * of the laws aforesaid ? * Then the savage must
be aware that he is misapplying these laws; for taboo is for him
a mystic aflFair. But if he knows he is indulging in error, why

' Dr. Frazer write©, Ledures on t)i4i Eadtj History of tlie Kingships 53, * It is
not a taboo to any, '^Do not put your hand id the fire " ; it 18 a rule of common
sense, because the forbidden action entails a real, not an tmaginary^ eviL' It is
not a taboo, but a rule of common prudence; for the savage. But not for the reason
alleged. In his eyes there Is nothing imaginary, but something terribly real, about
the death or other disaster he observes to overtake the taboo-breaker. How, then^
does he come to bring this kind of evil under a category of its own ? Surely it
ought to be the prime concern of Anthropology to tell us that

IS TABOO A NEGATIVE MAGIC?

225

does he not mend his ways ? Clearly Di\ Frazer caanot mean his
explanation of magic or of taboo to be an explanation of what it is
for the savage. Now, perhaps he is entitled to say that magic, in
his sense, is not a savage concept or institution at all^ but merely
a counter for the use of the psychology that seeks to explain the
primitive mind not from within but from without. He is, however,
certainly not entitled to say that taboo is not a savage concept or
institution. In Polynesia tapii is a well-recognized term that sen es
as perhaps the chief nucleus of embryonic reflection with regard to
mystic matters of all kinds ; in some of the islands the name
stands for the whole system of religion.^ Moreover, from every
quarter of the primitive world we get expressions that bear the
closest analogy to this word. How then are we to be content with
an explanation of taboo that does not pretend to render its sense as
it has sense for those who both practise it and make it a rallying-
point for their thought on mystic matters ? As well say that taboo
is ' superstition ' as that it is ' magic ' in Dn Pi-azer s sense of the
word. We ask to understand it, and we are merely bidden to
despise it.

If, on the other hand, we cast about amongst genuine primitive
notions for such as may with relative appropriateness be deemed
equivalent to the idea of magic, as that idea is to be understood and
employed by a psychology that tries to establish community between
savage and civilized thought, we have the choice between two
alternative types.

My own preference is for those primitive expressions that are
definitely dyslogistic or condemnatoiy, as when we speak of the
' black art *. The clearest cases that I know are AustraUan. Thus
the arungquiUha of the Aiiinta is * associated at bottom with the
possession of supernatural evil power'." Perhaps we may say
broadly that, as contrasted with churinga^ the term stands for magic
as opposed to rehgion— for magic, that is, as the witch-haunted
England of the seventeenth century understood it, namely as some-
thing anti-social and wholly bad. The Kaitish ittha seems to be the
exact analogue of arungtpiiltha '^ ; and so do the mupam of the Yerkla-
milling,* the 7nung of the Wurunjerri,^ and the guhhurm of the

* Cf. E. Tregear, 3Iaori 'Polynesian Comparative Dictionary^ s-v. /apw.
' Spencer and Gillen, op. cit.^ 548 n.

* Spencer and GiUen, The Northern Tribes of Central Australia^ 464 n,

* A W. Howitt, The Naiive Tribes of South-East Australia^ 450.
' Op, cit,, 365,

18 TABOO A HEGAUVE MAGIC?

Tqiiu' In all these eaeee the noCioii aciftitfT to be that of a wonder*
wofktng of a fompletoly n ojua t m kmd. Amnngwt the Antnta
a man eaof^ prartwing anch magic ia aemerrilj pmtdied, and
imbabfy killed'

Some, boweirer, might cho oa a mther to aoBgn tte meaaing of
^magie' to the wonder^worldng in geoecal, and not simplj to ita
h$d flriety. Ihn9 amongat the lart^nentioDed Tain ' eril magic '
msf be practieed by the gtmmeru or medicine man ; bat in this tribe
he 18 the leader of society, and a wielder of good ^ipematural power
no leas than of evO. The wonder-working power he poaaeasee goes
by the comprehenaiTe name of jaia, translated ' magic ' by Howitt^
and deaeribed as an ^ immaterial force ' set in motion not only by
tbBgommera but also by certain sacred animals.- Here we seem to
haye a case of that yefy wideepiead notion of which the most
&mous representatives are the mafia of the Pacific and the orenda of
the Iroquoia. A good deal of attention has lately been paid by
anthropologists to these latter expressions, and I may pertiaps be
permitted to take certain of their findings for granted. It would
appear that the rootridea is that of power — a power manifasted in
sheer luck, no doubt, as well as in cunning, yet, on the whole,
tending to be conceived as a psychic energy, almost, in fact, as what
we would call * will-power '.* Further, though it may be that every
being poes o BB oe its modicum of mana^ the tendency is for the word
to express extraordinary power, in short a wonder-working.

Now between the ordinary and the extraordinarj-, the work-
a-day and the wonderful is a diiference, if you ^vill, of degree rather
than of kiuA The sphere of the miraculous is, subjectively, just the
sphere of a startled experience, and clearly there are endless degrees
in the intensity of felt surprise ; though society tends to fix hard-
and-fast limits within which surpiise is, so to speak, expected of
one* How the savage proceeds to differentiate the normal from the

' Op. cU,, 872,

• Spencer and Gillen, Native Tribes, 536,

• C^ cU., 688, bm-h

• It IB very interesting to note, as Tregear^s excellent dictionary, a.v, mana^
illiblef) one to do at a glaQee, how the root mana underlies an immenae number of
the terms by which psychical faculties xuid states are rendered. Thus in Samoan
we fiod mana*o to desire^ wish, nianatu to think^ manamea to love, atuamanatu to
have a good memory ; in Tahitian matioo to think, manavaru eager desire ; in
Hftwaiian nianao to think, manafim thought, manaoto to believe, manaiva feelings^
affections ; and bo on.

IS TABOO A NEGATIVE MAGIC?

227

abnormal was brought home to me in the course of an infcei'view
I was accorded by the Pygmy * chief Bokane.^ I was trying to
verify CoL Harrison's statement - that if a Pygmy dies suddenly the
body is cut in two to see wliether or not the death is caused by
oudah — the 'devil', as Col, Harrison renders it, though, for my part,
I could not discover the sKghtest trace of personality attacliing to
this evil principle. ' I asked Bokane liow his people told whether
the death was due to omlah or not. He replied that, if an arrow-
head or a large thorn were found inside the body, it was an arrow
or a thorn that had killed the man ; but, if nothing could be found,
then ouduh must have done it If a dangerous animal kiOed a man,
I learnt on further inquiry, it was not o?^rfa/?, but it was ouduh if you
cut your finger accidentally. When strange sounds were heard in
the forest at night and the dogs howled, that was oudah. On some
such lines as these, then, we may suppose other savages also to Iiave
succeeded in placing the strange and unaccountable imder a category
of its own. In the case of mujia and orenda 1 am inclined to think
that the core of the notion is provided by the w^onderfol feats —
wonderful to himself, no doubt, as well as to his audience — of the
human magician ; which notion is then extended to cover wonder-
working animals, nature-powers, and the like by an anthropomorphism
which is specificiilly a * magomorphism ', so to say. Of course other
elements beside that of sheer sui-prise at the unusual enter into the
composition of a predominant notion such as that of mana^ which in
virtue of its very predominance is sure to attract and attach to itself
all manner of meanings floating in its neighbourhood. For example,
as the history of the word * mystic ' reminds us, the wonderful and
the secret or esoteric tend to form one idea. The Australian wonder-
worker owes no little of his influence over the minds of his fellows
to the fact that in most tribes an exhibition of his power forms part
and parcel of the impressive mystery of initiation. Let it suffice,
however, for our present purpose to identify mana with a wonder-
working power such as that of the magician — a power that may
manifest itself in actions of the sympathetic type, but is not limited
to this type, being all that for the primitive mind is, or promises to

* I spent about five hours in all in private talk with the Pygmies, assisted,
1 need hardly say, by an interpreter, at Olympia in London, Jan. 8 and % 1907,

^ Life aniont^ the Ff/gmks, Lond. 1906, 20.

^ Nothing, apparently, is done to avert or propitiate oudah, Bokane denied
that the pots of honey placed at the foot of trees were for omhih.

<J2

228

IS TABOO A negatht: magic?

be, extraordinarily effective in the way of the exertion of personal,
or seemingly personal, will -force.

Now, if * magic ' is to mean 7nana (which, however, is not Dr.
Frazer's sense of * magic ', nor, indeed, mme, since I prefer to give it
the uniformly bad meaning of artingqiiiltha^ that is of the anti-social
variety of mavm\ then in describing taboo as negative magic we shall
not, I believe, be far wide of the mark. Taboo I take to be a mystic
affair. To break a taboo is to set m motion against oneself mystic
wonder-working power in one form or another. It may be of the
wholly bad variety. Thus it is taboo for the headman of the water-
totem in the Kaitinh tribe to touch a pointing-stick lest the ' evil
magic' in it turn all the water bad.^ On the other hand, many
tabooed things, womaivs blood or the king's touch, have power to
cure no less than to kill ; whilst an almost wholly beneficent power
such as the clan-totem or the personal manitou is nevertheless taboo.-
Indeed, it is inevitable that, whenever society prescribes a taboo in
regard to some object in particular, that object sliould tend to assume
a certain measure of respectabihty as an institution, a part of the
social creed ; and, as the law upholds it, so it will sui'ely seem in
the end to uphold the law by punishing its infraction. It is to
be remarked, however, that many taboos prescribed by the
primitive society have regard to no object in particulm% but are of
the nature of general precautions against mystic perils all and
sundry, the vaporous shapes conjured up by unreasoning panic. It
is instructive in this context to consult the admirable account given
by Mr. Hodson of the communal taboos or yenrms observed through*
out the Manipur region.^ On all sorts of occasions the gennahura or
reUgious head of the village ordains that the community shall keep
a (fcnna. The village gates are closed, and the friend outside must
stay there, whilst the stranger who is within remains* The men
cook and eat apart from the women dming this time. The food
taboos aie strictly enforced.* All trade, all fishing, all hunting, all

' Spencer and Gillen, Nortltern Tribes^ 4G3.

* Is Dr. Frazer henceforth prepared to explain totemism on purely sjinpa-
thetic principles? It would, on the other hand, he easy to ahovv that the ideas
of mana and of manUot$ and the Uke go very closely together.

* T. C. Hodson, 'The *'Genna" amongst the tribes of Aj^am/t/* A. /., xxxvi.
»2aqq.

* Some of these food taboos have a sympathetic chaiucter. Thus 'youn^
unmarried girls are not allowed to taste the flesh of the male of any animal or of
female animals which have been killed while with young \ ib. 98. Even here.

IS TABOO A NEGATIVE MAGIC?

229

cutting of grass and felling of trees are forbidden. And why these
precautions? Sometimes a definite visitation will have occurred*
* Phenomena such as earthquakes and ecHpses, or the destruction of
a village by fire, occasion general (jemms, . • , We also find general
gennas occasioned by the death of a man from wounds inflicted by an
enemy or by a wild animal^ by tlie death of a man from snakebite
or from cholera or small-pox, or by tlie death of a woman in child-
birth/ ^ At other times nothing untoward has happened, but some-
thing important and ' ticklish * has to be done — the crops sown, the
ghosts laid of those who have died during the yeai\ It is a moment
of crisis, and the tribal nei^ves are on the stretch. Mr, Hodson,
indeed, expressly notes that ' the effect of qennus is certainly to pro-
duce in those engaged in them a tension which is of great psycho-
logiciil interest'.^ Is not what he takes for the effect rather the
caiLse of gennas? Anxiety says *Let us abstain from all acts that
may bring upon us the ill-will of the powei's \ Anxiety sees every
outlet of activity blocked by a dim shaj^e, endowed with no definite
attributes such as the sympathetic theory is obliged to postulate, but
tationed there as simply a nameless representative of the environing
XTnknown with its quite unlimited power of bringing the tribal manu —
its luck and eumiing — to nought by an output of superior numa^ to
be manifested who knows how?

It may be objected that, whereas we have made it of the very
essence of mana that it should be indefinite and mysterious in its
effects, there can be nothing indefinite or mysterious on the Dyak
view — to recur to the example fiom which we started^ — about the
effect of deer-meat, since it produces timidity exactly as it might be
tliought to produce indigestion. Perhaps it is enough to reply that
to the savage a fit of indigestion would likewise be a phenomenon
explicable only in mystic terms. The common sense of the primitive
man may-— to take Dn Frazer's instance — recognize that normally
and as a matter of course the fire burns whoever thrusts his fingers
into it ; but the moment that the fire burns some one * accidentally ',

however, an element of uiiracle entei-s, unless the Manipuria find pailhenogeneaia
no mare odd than the Arunta are liy some supposed to do. Another taboo is
on dog's flesh, the mygtjc penalty being an eruption of boils. Here there is no
obvious sympathetic connexion. Boils ai^ uncanny, and have to be accounted for
on mystic lines— if not synipiithetically, yet by some reference to evil magic ; for
tlisea'^e is always evil magic for the savage j cf. Sf>encer and GOlen, Native
Tribes, 548.

' ih. m. * iK 101,

230

IS TABOO A NEGATIVE MAGIC?

aa we say, the savage niiiid scents a mysteiy. Just so for the
Pygmy, His knife acts normally so long as it serves him to trim liis
arrow-shaft. As soon, however, as it sUps and cuts his hand, thei-e
is oudah in, or at the back of, the * cussed * thing. Given, then, anj'-
thing that behaves * cussedly ' with regularity, that is nomially ab-
normal in its eflfects, so to speak, and a taboo or customary avoidance
will be instituted* It becomes the duty of society to its members to
keep before their eyes the nature of the direful consequences attend-
ing violation of the rule. Society shakes its head solemnly at care-
less youth, and muttei-s fiopiiat. Careless youth does not believe all
it is told, yet is nevertheless impressed and, on the whole, abstains.
Kafir children must not eat certain small birds.' If they catch
them on the ^'eld, they must take them to their grandparents, who
alone may eat the body, though the childi-en are given back the
head, ' If the parents catch children eating birds on the veld, they
tell them they will turn out witches or wizards when they grow up/
Here we have the mystic sanctioru And there is a social sanction
in reserve. *The boys naturally get sound thrashings from their
fathei^, who feel it their duty to prevent their sons from turning out
abandoned wretches in after life.* Nevertheless, youth is sceptical,
or at any rate intractable, * Children do not see the logic of this
rule, and consequently tiy to eat the bird on the veld, when they
think they will not be found out. . , • There is no time when boys
and girls are so fi ee from observation as when watching the fields ;
consequently, at such times they have glorious feasts off the birds
they catch/ Now the spiipathetic principle may underlie this
food taboo, or it may not, but clearly by itself it is not enough to
account for the customaiy obser\^ance in the concrete. Society has
to keep the taboo going, so to say ; and to keep it going it relies
partly on the vis a tenjo of brute force, but still more on the sug-
gestion of mystic evil in store for the oflfender, not an imaginary
evil, pace Dr. Frazer, but what is quite another thing, an evil that
appeals to the imagination, an indefinite, unmeasured, pregnant evil,
a visitation, a doom, a judgement.

Hitherto we have had in view mainly such cases of taboo as
seemed most closely bound up with the sympathetic principle, minor
matters of routine for the most part, outlying and relatively isolated
portions of the social system, which for that reason might be ex-

' Dudley Kidd, Savage Childhood, a Studtf of Kafir Children, 193.

IS TABOO A NEGATIVE MAGIC?

231

p€ct€d to contain their own raison d'etre unaffected by the trans-
forming influence of any higlier synthesis. If, however, we turn to
the major taboos of primitive society, the classical wellnigh universal
cases of the woman shunned, the stranger banned, the divine chief
isolated, and so on, how infinitely more difficult does it become to con-
ceive sympathy, and sympathy only, as the continuously, or even the
originally, efficient cause of the avoidance. Unfortunately, considera-
tions of space utterly proliibit a detailed treatment of matters covering
so wide an area both of fact and of hj^othesis. It must suffice here to
assert that the principles already laid down will be found to apply to
these major taboos with even greater cogency. Here, too, there are
at work both a social and a mystic sanction (so far as these can be
kept apart in thought, the mystic sanction being but the voice of
society uttering bodings instead of threats). As for the mystic
sanction, we shall probably not be far wrong if we say that the
woman has manay the stranger has 7uaua^ the divine chief has numa^
and for that reason pre-eminently are one and all taboo for those who
have the best right to determine the moaning of taboo, namely those
who practise and observe it.

If there were room left in which to consider these taboos in some
detail — the three notable cases mentioned do not, of course, by any
means complete the list of taboos of the first rank* — it might turn out
that in oin* running fight with the upholders of the sympathetic theory
serious opposition must be encounteied at certain points, yet never
so serious, let us hope, that it might not be eventually overcome.

Thus the first case on our list-— that of the taboo on woman —
provides our opponents with a really excellent chance of defending
their position. There ciin be no doubt that a sympathetic interpreta-
tion is often put upon this taboo by savages themselves, Mr. Crawley,
who has made tlie subject of what he tenns the sexual taboo
peculiarly his own, brings forward evidence that, to my mind at
least, is conclusive on this point- Among the Barea man and wife
seldom share the same bed, the reason they give being that * the
breath of the wife weakens her husband'. Amongst the Omalias
if a boy plays with girls he is dubbed * hemiaphrodite ', In the

* Thus one of the most notable and widespread of talx>os is that on the dead,
Sympathetic interpretations of this taboo are by no means unknown amongst
savages, but it would not be hard to show that they do not exhaust the mystery of
death, of all human concepts the most thickly emiTapiied in imaginative at mosplierB.

* E* Crawley, The Mifstic Eos€j 93, cf. 207 sqq.

232

IS TABOO A NEGATIVE MAGIC?

Wiraijuri tribe boys are reproved for playing with girls, and the
culprit is taken aside by an old man, who solemnly extracts from
his legs some * strands of the woman's apron * -which have got in*
And so on in ca% after case. Here clearly what is piimarily feared
is the transmission of womanly characteristics, in a word, of
effeminacy. Mr. Crawley even goes so far as to speak of the behef
in such transmission as Hhe chief factor in sexual taboo 'J Whether
this be so or not,^ he Ukewise shows, with singular clearness and
force, that it is not the only factor. Owing, he thinks, to a natural
nervousness that one sex feels towards the other, as well as to the
unaccountable nature of various phenomena in the life*history of
woman such as menstioiation and child-birth, the notion of her as
simply the weaker vessel *is merged in another conception of
woman as a '* mysterious" pei'son . . . She is more or less of a
potential witch '.-^ With this I cordially agree, and shall not labour
tlie point more except to the extent of asking the question, How, on
tlie hypothesis that what is dreaded is simply the transmission of
womanliness, are we to account for the fact — to quote but the best-
known story of tlie kind — that when an Australian black-fellow
discovered his wife to have lain on his blanket he wholly succumbed
to terror and was dead within a tbrtnight?^ Only a twilight fear,
a measureless hoiTor, could thus kilL And to show how mixed
a mode of thought prevails as to the workings of the sanction set
in motion, in a very similar case from Assam it is not the man but
the woman who dies of fright.^

The case of the taboo on strangers seems at first sight to affoiHi
a clear proof of the effect of mere strangeness in exciting dread,
especially when we compare the results of contact with novelties
of all kinds. Dr. Jevons, however, argues that ' strangers are not
inherently taboo, but, as belonging to strange gods, bring with them
strange supernatural influences'.^ In support of this view he

» ib. 207.

* Mr. Crawley does not tell ua on what principle he woulil proceed in estimate
predominaQce as between such factors. 1 should have thought that the moral
of hiB excellent study, aWunding aa it does in psychological inaight, was to lay
stress on the subconscious grounds of action rather than on the reasons whereby
more or less ex post fach the dawning reflection of the savage seeks to intei*pi'et
and justify that action. I mjrself believe the sympathetic explanation to be little
more than such an ex post facto justification of a mystic avoidance already in
fiill swing.

' iK 206. • Jmm Antlu Inst,, ix. 4&8. * Hodson, op. cit., 100.

* An Introduction to the Histot^i/ of Etligitm, 71.

IS TABOO A NEGATIVE MAGIC?

233

instances the fact that new-comers are frequently fumigated to
drive away the evil influences they bear in their train. But, after
all, there are no taboos that reUgion has not learnt to neutralize by
means of one or another ceremonial device* Woman, for example,
is inherently taboo, yet with proper precautions she may be married.^
So too, then, strangei's may be entertained after a purifying ceremony.
It by no means follows^ however, that they have lost all their mystic
virtue, any more than it follows that woman has ceased to be mys-
terious after the marriage ceremony* Witness the power to bless
or to curse retained by the stranger within the gate — a matter for
the first time brought clearly to light by Dr. Westemiarck's striking
investigation of the rehgious basis of primitive hospitality.- Mean-
while, even if Dr. Jevons'a contention were to be granted that the
taboo on strangers is really a taboo on the tabooed things he may
have been in contact with, it is hard to see how the sympathetic
explanation of taboo is going to be stretched to cover the indefinite
possibihty of definite sympathetic contagions of all sorts. We are
left asking why mere uncertainty in itself can rouse imaginative
fears — ^a line of inquiry that must presently lead to the conclusion
that mere strangeness in itself can do the same.

The third of our cases— that of the tabooed chief — need not
detain us long. At all events in Polynesia, the eponymous home of
taboo, they have no doubt about the explanation. The chief has
Hmmi^ and therefore he is feared. Men do not dread contact with
the king lest they become kingly, but lest they be blasted by the
superman's supermanliness. Such, at least, is the native theoiy of
the kingly taboo on its religious side. On its higlily developed
social side it is a fear of the strong arm of the State mingled with
a respect for established authority— just as religious taboo is for the
most part, not all cringing terror, but rather an awe as towards
mystic powers recognized by society and as such tending to be
reputable.

We have cast but a rapid glance over an immense subject. We
have but dipped here and there almost at random amongst the end-

^ I accept Mr. Crawley's hypothesis that ' marriage ceremonies neutralize the
dangers attaching to union between the sexes '. Tfte Mystic BosCy 322.

* E. Weatermarck, The Orttjin and Bcvel<yptn€ni of the Moral Ideas, i. 583 sqq.
Pr. Westermarck's view, by the way^ is that ' the unkno\^Ti stranger, like every-
thing unknown and everything strange, arouses a feeling of mysterious awe in
superstitious minds '.

284 IS TABOO A NEGATIVE MAGIC?

less facts bearing on our theme to see if the sympathetic principle —
a perfectly genuine thing in its way — ^would take us to the bottom of
the taboo feeUng and idea. We conclude provisionally that it will
not Indefinite rather than definite consequences appear to be
associated with the violation of a taboo, and that because what is
dreaded is essentially a mysterious power, something arbitrary and
unaccountable in its modes of action. Is, then, taboo a n^ative
mana'i Yes — ^if mana be somewhat liberally interpreted Is it a
negative magic, understanding by magic EPfrmpathetic action ? With
all my respect and admiration for the great authority who has pn>
pounded the hypothesis, I must venture to answer — No.

THE ETHNOLOGICAL STUDY OF MUSIC

By CHARLES S. MYERS, M.A., M.D.
Professor of Psych olooy, King's College, London

Intmdndory. ~lt is hardly surprising that the ethnological study
of music lias been neglected until recent times. For, in the first
place, its recognition as a serious branch of scientific investigation
has been veiy tardy. Not long ago we were all of us apt to look on
primitive music just as the Greeks regarded the language of their
neighbours. We now know that, disorderly and meaningless as
unfamiUar language and unfamiliar music at fii*st appear, an in-
herent order and a meaning are I'evealed after suflBcient study and
habituation.

In the second place, the subject ha^ until recently demanded
a worker who is alike a trained ethnologist and a musician. But
within the last fifteen years, the use of the phonograph has
enormously lightened the weight of musical knowledge, which the
worker must otherwise take with him into the field. It is now
possible for the ethnologist of very moderate musical attainments
to collect phonographic records and other data of great musical
interest which can be worked out by the specialist at home.

Tlie theorist should henceforth have no cause for complaint of
paucity of materiah Nor need he longer rely on the unverifiable
guesses and errors made by independent and often untrained
obser\ers in the field,

Cmitaminafimi of primitive music, — There is hardly any other
branch of etlmology where so much remahis to bo done, and wliere
the opportunities for research are so rapidly vanishing beyond recall,
as the study of comparative music. The borrowing and adulteration
of music prooeed apace. When tribes, formerly hostile, become
pacified, fresh routes are opened up for the mutual exchange and
contamination of diflferent styles of primitive music. When sacred
and profane European tunes are introduced by the missionary or the
trader, unpolluted aboiiginal music soon has a precarious existence.

236

THE ETHNOLOGICAL STUDY OF MUSIC

We have evidencje of these conditions in the influence of Arabic
or Portuguese tunes introduced into Africa, and in the spread
of favourite native airs throughout North America and throughout
Australia,

The effectiveness of a borrowed tune amply compensates for
the strangeness of the words that may belong to it. The words are
commonly sacrificed to the tune. So long as the latter is acceptable,
it mattei-s little that the former are meaningless. We frequently
find that liberties are taken with woixis, or that meaningless words
or syllables are introduced into primitive music* Yet another cause
of the presence of meaningless words lies in the antiquity of the
music, Tlie words become so archaic, or their sense was originally
so involved or so symbolical that all meaning gradually disappears as
the song is handed down from generation to generation.

The expressive fund ion of mum, — Music is a recognized means
of intercommunication, and must hence be regarded as a language.
But the language of music diffei's from verbal language in that it
can communicate only emotions (or feeUngsj, while verbal language
serves for cognitive (intellectual) as well as for emotional expression,
Tims wlien we employ words^ we commimieate not merely a feeling,
say of joy, anger or sorrow, but iilsn the events or ideas which are
bound up in those feelings.

In expressing feelings by spoken language, our words vary in
pitch according as we are niixking a statement or asking a question,
our voice changes in loudness according as we are angry or calm,
the timbre of our voice differs according as we are sarcastic or
persuasive, the speed of our words and the rate of respiration alter
according as we are excited or depi-essed. Now music, as we have
said, can only eomminiieate such states and changes of feeling. Music
can awaken in us feelings of joy, excitement, sadness, resignation,
courage, uncertainty and the like, but it cjmnot communicate to us
the ideas wliich are the cause of such feelings. These ideas are the
product of each hearers fancy. That is to say, the language of
music is devoid of acknowledged signs for cofpiitive expression.

To investigate the degree of universality of those signs in
verbal and masical language which serve to communicate states of
feeUng, is a matter of no small interest. We would know how far
the modifications of verbal language in respect of pitch, loudness,
timbre and tempo sci-ve as universal methods of communication,
and similarly how far the feelings of sorrow, joy and tlie like, wliich

THE ETHNOLOGIC.iL STUDY OP MUSIC

237

a given piece of music evokes in the community that produced it, ai^
shared by the niembem of other communities more or less advanced
in civilization.

The origin of mime. — We may regard musical and verbal
language as derived from a common source^ namely from the
tendency to give vent to feelings by vocal expression. There are,
however, other theories as to the origin of music which lay stress
on more special factoi^s. One of the objects of the ethnological study
of music should be the determination of the importance of these
various factors.

It has been suggested, foi- example, that music arose from the
imitation of notes of birds and other natural sounds — a conjecture
closely analogous to the supposed onomatopoeic origin of verbal
language. The suggestion has also been put forward that music
began when primitive man vied one witli the other in exliibiting
his superior attractiveness before women. Other theorists, looking
to the value of rhythmical music in finihering work and in dancing,
and having regard to the delight taken by primitive people in the
beats of the tom-tom, have laid chief stress on rliytlnn as the source
of all music. Here, again, is a conjecture which can only be verified
by the systematic study of piimitive music.

Rhythm and melody. — Wliile some examples of primitive music
are characteiized by a total absence of rhythm and appear to be
melodic elaborations of the recitatif, in other examples rhythms of
sucli complexity are introduced that they defy analysis by the
civilized European ear. Not infrequently the accents or measures
in the melody are opposed to those in the accompaniment. In
India and apparently among the Arabs and ceitaha other peoples,
successive notes of very different duration are grouped together and
recognized as a unit, each unit sometimes receiving a si>ecial name
and having a special use according to circumstances.

Rhythm and harmony.— The widespread occurrence of complex
rhythms among primitive peoples is perhaps intimately related to
their generally scant feeling for harmony. In Europe the develop-
ment of polyphony (in which various independent melodies are
sung simultaneously) was regulated by the growing regard for
consonances and dissonances. Certain tones when sounded together
appeared agreeable, othei-s were deemed unpleasant. Thus arose
the distinction between consonant and dissonant combinations.
The most perfect consonance is given by the octave, i,e. by two

238

THE ETHNOLOGICAL STUDY OF MUSIC

tones whose vibratioii-fi'equeiicies stand in tlie ratio 1 : 2. (The
number of vibrations per second determines the pitch of a tone,)
The next most perfect consonance is given by the fifth, the corre-
sponding ratio for which is 2 : 8, Then follows the fourth (3 : 4),
Our attitude towards the various musical intei-vals has differed con-
siderably at different stages in the histoi-y of European music. Thus
the thirds, major (4 : 5) and minor (5:6), and the sixths, major (3 : 6)
and minor (6 : 8), wliich ai'e often called imperfect consonances, were
not admitted as consonances until the thirteenth and fourteenth
centuries. At the present day, not all the imperfect consonances
are admitted into our system of harmonies. The liarmonic tritone,
for example (5 : 7), does not enter.

Fimofk — The degree of consonance is literally dependent on the
extent to wliich the two tones ^ sound together' (con-sonare). So
complete is the fusion between a tone and its octave that even the
most musically gifted people find difficulty in deciding whether two
such tones are simultaneously sounding or only one. Less musical
people make similar mistakes when the simultaneous tones employed
are separated by a fifth, or by another less consonant interval. The
less the degree of consonance, the less erroneous the decision,
Wlien the interval is distmctly dissonant, e.g. a major (8 : 9) or
minor (15 : 16) second^ or a major (8 : 16) seventh, there is practically
complete absence of fusion.

As polyphonic music began to develop in Europe, the growing
feeling for consonance and dissonance demanded that a strict
uniform tempo be kept by the various executants. Clearly the
harmonious eflfects would be utterly spoiled miless the parts were
in exact time with one another. This condition was most easily
attained when the accents recurred regulai*ly and the rhythm pre-
served a fairly simple chai^acter.

Polyphony. — Polyphonic music is far commoner than is generally
supposed m the music of other than Eiu^opean peoples. The pur-
poseful use of simultaneous harmonies, especially of octaves and
fifths, is not unusual among semi-civilized people. But tlie several
parts are invariably permitted a freedom of movement which is
denied to our own music, and the diflferent simultaneous rhythms
are allowed full scope for independent development. Such poly-
phonic music — or to adopt a more appropriate name that has been
suggested, such ' heterophonic ' music — surely demands of the native
audience the same oscillations of attention as occur in us when we

THE ETHNOLOGICAL STUDY OF MUSIC

239

listen to two persons talking simultaneously. Our attention turns
alternately now to the one voice, now to the other, and we inten-
tionally neglect the jarring eflfect of the snnultaneous voices upon
consciousness. For our comprehension of the sense, it mattera
tittle what times these speakers keep relatively to one another. In
like manner it is not considered imperative for the individual parts
of a primitive orchestra to keep a preKcribed time. Variations are
permissible, dictated by the taste of the perfoi-mers.

Hamiony in primitive mtmc. — To what stage and by what steps
the feeling for harmony has advanced among primitive people, can
only be settled by systematic investigation. Attempts have been
made to ascertain whether such people show a preference for con-
sonant and a dislike for dissonant pairs of simultaneous tones, or
whether they regard various pairs or triads of tones as differing in
affective (e,g, exhilarating or depressing) value* But no satisfactory
results have yet been obtained* It is cleai' that both native and
European intervals should be presented, that the interv^als should
be sounded on native and not merely on European uistruments,
and that repeated judgements must be obtained before reliance can
be placed on such comparisons. Other investigatoi-s have hoped
to arrive at an answer by playing primitive melodies on the
piano, harmonizing them now in one way, now in the other.
But the likes, dislikes or indifference of the natives, ascertauied
by such a rough method of expeiiment, cannot be accepted as
trustworthy.

We must bear in mind that the disorderly use of simultaneous
tones in primitive orchestra or chonis does not necessarily imply an
inability to distinguish between harmony and discord. One may be
quite able to discriminate between two experiences, although in
practice one may totally neglect the differences between them ; we
may, for example, give the same name to two ideally distinguishable
objects. To argue that primitive man cannot distinguisli blue from
green (or salt from sweet) because he designates them by the same
name would be absurd. But it is haidly less absurd to insist that
the feeling for consonant intervals is absent among a given primitive
people which totally disi-egards it in then* music. It is quite con*
ceivable that the neglect of the principles of hannony in primitive
music may be due partly to the difficulties of securing exact intona-
tion, partly to the peculiar mtei^v^ils and scales which have become
imposed upon them, and partly to such an imcontrolled desire for

240

THE ETHNOLOGICAL STUDY OF MUSIC

massiveness of souud that any tendencies for putting their feeling
for harmony into practice are at once repressed.

Styles and social futwHon of viusic. — Some of the changes which
the music of a given people has undergone in the course of its
development may be revealed by a careful comparison of the older
with tlie moi'e modern tunes. Nearly every people, however
primitive, preserves what we may term its classical miLsic. Such
music often becomes invested with a sacred character. It may be
performed only in secret initiation ceremonies, or during religious
observances. In this comiexion the native m3^hs regarding the
origins of music and musical composition should be studied,

Tlie position of music within a community is no doubt largely
responsible for the number of coexisting styles of music, and for
the degi'eo of conservatism obtaining. When musical instruments
are to be found in nearly eveiy family as in Japan, there are many
diflferent styles of music, which are strictly confined to certain
classes of performers. Where instrumental music is limited to
professional players, its theory and practice are apt to be treated as
secret, and ai*e regiu'dcd as the property of the guild. Under such
circumstances musical education is dependent solely on tradition,
and any attempt at musical notation is discouraged. In Japan the
beginnings of notation are to be seen, but the figures themiii
employed refer to particular instruments, e.g. to the hole which has
to be unstopped on the flute, or to the fret on the guitar at which the
finger has to be placed in order to produce the required note. Musical
notation also exists in India and China, and was employed by the
ancient Greeks and mediaeval Arabs.

Scuhis,~TonoH may be regarded as the vocabulary of music.
If we collect all the tones which a given people ever introduce into
their music and arrange these tones as a scale in the order of their
pitch, such a collection is analogous to a vocabulary of words. But
just as not every word which is to be found in a dictionary is
appropriate in a given hterary work, so not every note which
occui's in such a scale can be indiscriminately employed in a given
piece of music.

We have thus to distinguish two kinds of scales, of which the
one is obtained as described above, by collecting all the tones
utilized in the various tunes of a given people, while the other is
formed by collecting tones which are to be found in a single tmie.
We may term the former a ' general ' scale and each of the latter

THE ETHNOLOGICAL STUDY OF MUSIC

241

a * particular ' scale. The music of a given people, therefore, consists
of a single general scale^ and of a series of particular scales.

Now the tones wliich a people employ in their music are not
merely dependent on their aesthetic appropriateness. Their exact
pitch is in part determined by the construction of musical instruments,
and by the difficulties of instrumental technique. The arrangement
of the holes of flutes and of other wind instruments is sometimes
dictated not by auditory considerations, but by principles of sym-
metry or by other deteiminants.

Moreover, mathematical principles have always influenced the
fixation of the pitch of tones, wherever civilization has sufficiently
advanced to enable calculation to do so. Among the Greeks,
Pythagoras divided the string in the ratio of 2 : 3, and the Chinese
shortened the pipe in the same ratio. They thus produced the
interval of a fifth, and they divided the shortened pipe or string
again in the same ratio, and repeated the procedure, thus obtaining
a geometrical progression of fifths, bearing these relations to the
initial tone : —

I (!)\ (in (i)s &c.

Other mathematical principles have also played a part ; so, too,
has the mystic value of certain numbers. It remains yet to be
proved by accurate observation how far many of the abstruse specu-
lations of the mathematical theorists are actually embodied in
practical music. We may hopefully look to discovering the ^ natural '
intervals that are employed by a given people when these control-
hng influences of authority and convenience are, so far as possible,
experimentally removed.

We must be on our guard against placing too great a reliance
on the speculations of comparatively modern theorists. In the
development of European scales, for example, it would be rash to
suppose that the octave has always been the distance theoretically
subjected to division. The earliest Greek melodies, for example,
appear to have had a much nan'ower compass. The tetrachord is
thought to have been the first attempt at a scale in Greece. It con-
sisted of the intei-val of a fourth divided into three parts. Another
added tetrachord subsequently completed the octave.

The mode of construction of the particular scales is found to
vary widely among diflferent peoples, and even among the same
people at diflferent times and in different kinds of music. Most
usually the octave of the particular scales is divided into five or

TTLOft B

242

THE ETHNOLOGICAL STUDY OF MUSIC

seven tones. The five-toned or pentatonic scale occurs in every
continent. A common form of it, found, for example, in Chinese,
Japanese and Scottish music^ omits the fourth and the seventh, so
that the octave starting from c runs thus : —

c d e g a Cy the intervals of which correspond to
1 1 1| 1 1| of our whole tones. Our
own heptatonic scale runs : —

cdefgabc^
1 1 I 1 1 1 I
of which the intervals between e and / and between b and c are
(approximately) half the size of the other intervals.

It is not unusual to find slight deviations from an otherwise
strictly pentatonic or heptatonic scale. A given tune may be ob-
viously pentatonic or heptatonic in structure, save for the inclusion
of one or two comparatively unimportant or ' grace * notes.

Many other forms of the pentatonic scale besides the above
are described, of the derivation and interrelation of which we are
wholly ignorant. Thus in Japan the following forms (and others)
appear to be in use :—

Many forms of the heptatonic scale were recognized in early
European music, but in modern times they have become limited
to two, the major and the minor. Among the ancient Greeks, for
example, we find the following modes ; —

(a) the Lydian :

c d e fff a b Cj

corresponding to our major mode.

THE ETHNOLOGICAL STUDY OF MUSIC

243

(6) the Ionic or Hypophrygian :

c d efg a}P c^

(d) the Aeolic :

cd^fga^d^Cj

corresponding to our minor mode.

(e) the Doric :
c^^fg(^^c,

{/) the Mixolydian :

{g) the Syntonolydian :

c d e P g a h c^

and further complications, derived from these, have been described.

These diflferent * modes ' are here written out, all beginning on
c merely in order to facilitate comparison. Apparently the Greeks
employed different modes according to the metre chosen, and they
came to attach broad distinguishing characteristics to each of the
modes. The Dorian mode, for example, was reputed to be severe
and virile, others to be smooth, erotic, suitable for boys, and so forth.
But writers differ so much in their attitude to the various modes
that it is impossible to lay much stress on their opinions. It is not
improbable that if we had more information as to the ways in which
different kinds of music are regarded by the theorists among modern
semi-civilized communities, some light might be thrown on the at
present obscure views held by the ancients.

In certain forms of Arabic music and on the Scottish bagpipe,
the following heptatonic scale occurs : —

gah-cde-fg,
showing intervals of 2 1| 1| 2 l\ 1| 2 semitones.

It will be observed that this scale gives a neutral third of three
and a half semitones, intermediate between the major and the
minor thirds.

Tlie origin of the pentatonic scale is unknown. It is easy to
conjecture that after the octave the fifth (e.g. c—g) may have been
the next recognized interval, and that by taking a fifth from c down-
wards — thus reaching / — ^and by raising the latter an octave, the
four tones c, /, </, c may have been reached. By such means the

R 2

244

THE ETHNOLOGICAL STUDY OF MUSIC

interval of a whole tone f—g would be also reached, and this interval,
once fixedj may have been imitated by placing a tone d at a similar
distance from c, and a tone a at a similar distance from g. Thus we
arrive at a pentatonic scale of the form c, rf, /, 9, a^ c.

Equal temperament — We have just hinted that small intervals
may be produced by judging equal distances. This seems actually
to have determined the formation of scales in Siamese and Javanese
music. The octave is here divided into seven and into five equal
intervals respectively.

But while we can only dimly conjecture the causes and methods
that have resulted in the construction of such * equally tempered '
intervalSj it is easy to understand the origin of the similarly tempered
general scale to which those of our own instruments that have fixed
tones, e,g. the piano or the harmonium, are attuned. Our particular
scales are almost always heptatonic, consisting approximately of five
whole-tone and two half-tone intervals. These intervals, as we have
said (p, 243), were once arranged in various orders, each order con-
stituting a mode. In modern times, however, our scales have be-
come restricted to two modes, the major and the minor, the most
important difference between which consists in the interval between
the first and the third tones of the scale. The major mode contains
the major third (4 : 6), the minor mode the minor third (5:6). The
intervals of our scale came at one time to be determined by the con-
sonant relation of tones to one another. Thus the distance c—g was
in the ratio 2 : 3, c— /in the ratio 3 : 4, and so on. But difficulties
at once presented themselves when the scale of a melody instead of be-
ginning on c, as in a previous melody, now began on dor e. A little
consideration of the new ratios involved will show tliat such changes
of key necessitate the construction of new intei-vals which are often
not quite identical with the tones of the previous scale. It was
in order to overcome this difficulty that the system of equally tem-
pered tuning, now in vogue, was introduced. The octave is divided
into twelve equal intervals of a semitone. None of the intervals,
whatever be the key of the major or minor mode, exactly corresponds
to the requirements of strict harmony. Every interval within the
octave is a compromise which is satisfactory in so far as it allows
us to employ a comparatively small series instead of an enormous
number of notes, in instruments like the piano which have fixed
tones.

It lias been shown that, in spite of such artificial mistxmings to

■

THE ETHNOLOGICAL STUDY OF MUSIC

245

which from infancy our ear is exposed, musical i>ersons still tend to
sing truly consonant, instead of tempered, intervals and to play con-
sonant intei-vals on instruments like the violin which have variable,
instead of fixed, tones. We have consequently to recognize * instru-
mental * scales, as well as the general and paiiicular scales with
which we have hitherto dealt. Further, we see how important it
is to study not only the instrumental but also the vocal music of
a given people,

Quurfer'tones.—The Arabian theorists included quarter- tones in
the scales which they constructed, and it has been stated that in
Syria a scale occurs consisting of equally tempered quarter-tones.
The various quarter- and third-tone scales described by Aimbian and
other writers are probably always general scales ; they are rarely, if
ever, particular or instrumental scales. Wlien quai*ter-tone intervals
occur in any piece of Arabic music, the notes concerned are only grace
notes or play an otherwise unimportant part in the melody.

In Indian melody, however, these grace notes are considered to
be of very great importance in adding to its expressiveness. Here
we find the so-called ' srutis \ intervals varying between one third and
one quarter of a tone, which are treated as essential features of the
melody* Yet these srutis never appear to be fixed by the frets of
the stringed instruments; they can be produced only by slightly
varying the tension or the position of the finger at the place of the
frets. Much inquiry is yet needed before the problems of Indian
music can be settled. At present we can only regard with con-
siderable suspicion the hitherto generally accepted view that in the
Indian general scale the octave, with its seven intei*vals, is sub-
divided into twenty-two tones. Recent investigators have sug-
gested that the Indian general scale is identiciil with our tempered
twelve-tone scale of chromatic semitones.

The ' ragas ' play a most important part in Indian music, but in
the face of so much disagreement among writers on the subject we
can form no clear idea as to what ragaa are. Certain of them are
deemed appropriate for certain seasons, some can be played only in
the day, othei-s by night. The rSgas ai-e symbolized as individuals,
male and female (raginis),

A raga is not to be identified with a scale, inasmuch as there are
several diflferent nlgas in the same scale. Nor is it synonymous with
the mode, as different ragas appear in the same mode. Yet every
raga is said to have a definite mode, and to obey the succession of

fi6 THE ETHNOLOGICAL STUDY OF MUSIC

int^TTals fotmd in a definite inia Mixed rtgw also occur, formed
from tbe ttnion of diffarent modes and sealea At present the whole
|MObl«li of IndUo muaie is imrolTed in obeenrity. Its solution re-
quires the eoK>peration of native and European mfifiinn, so that the
traditions and claims of the one may be Terified bj the accurate and
onbiaaBsd obaenratiocis of the other«

Harmome tiiterwi/j in melody. — In the tunes of very primitive
people, who always sing in unison and have no knowledge of poly-
phonic music, we often meet with aoeceasive tones which, if
sounded together, would produce true consonances. It has been
suggested that such consonances have been actually heard by these
peoplOt owing perhaps to their chance occurrence in nature or to the
occasional want of strict time when members of a chorus are singing
together. We may reasonably question whether such accidentally
occurring instances of fusion are req>onsible for the existence of
harmonic intervals in the melodies of very primitive folk who
never practise polyphonic music The appreciation of a relation
between consecutive tones is a far more plausible explanation, but
we are entirely ignorant of the psycholc^cal and physiological
basis of such appreciation.

So far as the smaller intervals are concerned, we have to bear in
mind that approximately whole-tone and semitone intervals (seconds)
are exceedingly common among such people, and that in folk music
generally the frequency with which the various intervals are used
decreases proportionately with their size. It is highly probable that
the smaller intervals have been determined rather by the feeling for
equal tone-distances than by any feeling for simultaneous harmony.
The feeling for tonality may also have helped in the definition of and
preference for the smaller intervals in melody.

Tonality. — By the feeling for tonality we mean the underlying
recognition of a tonic ; that is to say, a certain tone of a melody is
regarded bh the centre of gravity, to which all the other tones come
to have a felt reference and seek for the sake of restfiilness to return.
The tonic is not necessarily the lowest, nor need it be the final tone
of the melody. The feeling for tonaUty has developed pari passu
with the growing feeling for harmony ; but in a low degree it may
certainly exist independently of the latter.

Just as words are grouped into a single sentence and the
sentences are grouped into paragraphs, chaptens, and so on, so the
individual tones of music are grouped into a single section, sections

THE ETHNOLOGICAL STUDY OF MUSIC

247

into a single phrase, &c. The combination of such pmls into a
unitary whole has been greatly furthered by the feeling for tonahty,
the felt relation of the individual tones to a tonic. In the most
primitive music the feeling for tonality appears to be just dawning,
and here we find a corresponding minimum of unification and
method. It is as if the attention of primitive folk were incapable of
combining more than a few consecutive notes into a connected
whole. For a few seconds, perhaps, we catch a glimpse of tonality
and tonal relation, and then the tonic, or, as we should say, the key,
changes or maybe it is lost in the general chaos of disorder. More
definite traces of tonality have been met with in Chinese, Siamese,
and Japanese music, among several tribes of the North American
Indians, and in India where, it is stated, a special word, ama^ exists,
denoting the tonic.

Awareness of absolute pikL — Owing to the growing influence of
the feeling for harmony and tonality, we tend to judge of the
pitch of a tone in melody not absolutely but by its harmonic relation
to the tonic or to some preceding tone. Our attention is diverted
more and more completely from the absolute charactei's of a given
tone or tone-combination towards its relation as part of a larger
whole. Our musical education leads us to regard the interval as of
greater importance than the absolute pitch.

Nevertheless, in certain individuals, especially among the most
musically gifted, awareness of absolute pitch is unquestionably
present and may become developed among them to an astonishing
degree. A single note struck on the piano can be instantly named
and identified. If confused at all, the note is apt rather to be con-
fiised with its octave than with any neighbouring tone. The answer
is given as unreflectingly as if the subject were asked to name a pre-
sented colour. Each tone, Uke each shade of colour, comes to be
individually and absolutely recognized. Each immediately revives
itfi special name, a, a*, 6, &c. Some individuals excel best in giving
a name to a given tone, others in reproducing the appropriate tone
when the name is given.

We are entirely ignorant of the extent to which, and the frequency
with which, this awareness of absolute pitch occui-s among less
civilized and primitive peoples. Individuals in whom it is strongly
developed would naturally be averse to transposing a melody or series
of tones to another key. When once they had heard a tone or learned
a musical phrase, they would repeat it after a prescribed lapse of time

248

THE ETHNOLOGICAL STUDY OF MUSIC

(say, one minute or half an hour) in precisely the same pitch as that
in which they Iiad originally heard it Another method by which
a subject's awareness of absolute pitch could be tested (in which
memory is not involved) would be to put a native instrument into
his hands and to ask him at once to reproduce on it a tone which is
sung or played to him by the investigator. The success with which
the subject can directly accomplish this without any groping or
error would indicate the extent to which his awareness of absolute
pitch is developed. The constancy with which from day to day
instruments are attuned to the same pitch or songs are sung in the
same key would also serve as indications. Among ourselves the
transposition of a melody into the corresponding mode of another
key is accompanied by a decided change in aflFective value.

Conditions affecting apparent pitch, — ITie awareness of absolute
pitch has been shown to be closely dependent on the timbre of the
tones. Thus a person who succeeds perfectly well on one piano may
not succeed on a strange piano or on a different kind of instrument

The effect of timbre upon pitch is very striking, A sound rich
in overtones emitted, for instance, from a reed instrument appears
distinctly sliarper than one of the same pitch emitted from the Bute,
which is comparatively free from overtones, A loud sound is also
apt to be judged of higher pitch than a soft one. Care must there-
fore be taken in comparing the pitch of tones produced from different
instruments or with different intensities. For like reason it is of
interest to discover whether a native language has separate words for
denoting pitch, intensity, and timbre.

We have always to be on our guard against purely accidental
deviations from strict intonation. We may detect them by procuring
repeated phoDOgi*aphic versions of the melody at different times
from the same or different individuals. Deviations from exact
intonation are to be expected among primitive folk who are careless
and unmethodical in their artistic production generally. Such errors
naturally tend to be overlooked by the people in the absence of
any controlling feeling for hai*mony. Such errors are encouraged by
difficulties of technique, by temporary excitement, and by the various
feelings associated with the various tone-intervals. We ourselves,
for example, tend to exaggerate the difference between major and
minor thirds, making the former too large, and the latter too small.

Our attitu/ie towards stramje 7nnsic.~lt is esusy to see how a regai'd
for regular rhytlim, harmony and tonality, and the piinciple of

THE ETHNOLOGICAL STUDY OP MUSIC

249

equal temperament are responsible for the attitude of European
civilization towards music generally. No sooner do we hear a piece
of primitive or advanced music than we endeavour to interpret it in
terms with which custom has long familiarised us. Absolutely
without reflection we read into the music regular accents, we arrange
it in bars, we declare it to be in such and such a key, and to be in
the major or in the minor scale^ we identify its interv^als with those
of our own to which they most neai'ly correspond. We forget that
the complexities of rhythm may far exceed what we are accustomed
to, and that primitive music knows little of tonality, and nothing
of major or minor scale.

Thus it comes about that many examples of primitive music
are incomprehensible to us, just because they are not so readily
assimOated as those which are more nearly related to our previous
experiences. Our attention is continuously distracted, now by the
strange features and changes of rhythm, now by the extraordinary
colouring of strange instruments, now by the unwonted progression
and character of intei'vals. Consequently much familiarity is needed
before we can regard such music fi'om a standpoint that will allow
of faitliful description. We have first to disregard our well-trained
feelings towards consonances and dissonances. We have next to
banish to the margins of our field of consciousness certam aspects of
music, which, were it our own music, would occupy the very focus
of attention, TIius incomprehensibility will gradually give place to
meaning, and dislike to some interesting emotion.

Appendix

The maniptilatwn of the phonograph, — The principle of Edison's
phonograph is familiar to most people, A wax cyUnder rotating
about its horizontal axis is driven by clockwork (or by other
mechanism), A recording diaphragm, the * recorder *, is brought to
bear on the i^evolving cylinder. The recorder consists essentially of
a very thin glass disk, to the lower surface of which is cemented
a sappliire pointer or style. This sharp style cuts a shallow groove
on the wax cylinder. While the cylinder is revolving, the recorder
is so moved that it marks a continuous spiral groove from end to
end of the surface of the cylinder.

To obtain a phonographic record or * phonogram ', a blank
cylinder is placed on the phonograph, a trumpet is affixed to the

250

THE ETHNOLOGICAL STUDY OF MUSIC

recorder, the clockwork is started, the style of the recorder is
lowered on to the surface of the cylinder, and the desired sounds
are made to enter the trumpet. To reproduce the record, aU that is
done is to substitute the * reproducer ' for the recorder. The former,
like the latter, consists of a thin disk of glass, but to it is affixed
a blunt sapphire style which, when brought to bear on a phonogram,
follows the spiral groove and accurately reproduces the movements
previously made by the sharp style of the recorder. These movements
are communicated to the glass diaphragm of the reproducer and
transmitted to the external air as vibrations of sound, where they
are reinforced by means of the trumpet attached to the reproducer,

A complete outfit for ttiking and reproducing phonograms
consists of (1) phonograph (and accessories); (2) wax cylinders;
(3) recorder and reproducer j (4) spare parts in duplicate.

L The phonograph wliich I recommend to travellers m called
the Edison-Bell Standard Phonograph. It is enclosed in a well-made
box, and weighs 19 lb. It can thus be carried without difficulty.
I have heard surprisingly good records taken in the field with lighter
and cheaper phonogt^aphs, but I consider it dangerous to depend on
them, as the clockwork of such instruments is Uable to run
irregularly, and in other ways to wear badly.

The accessories comprise trumpet, oil-can, oil, and bmsh.

The same trumpet can quite well be used for recording and for
reproducing phonograms. In my experience the best form measures
about six inches in diameter at the mouth, and is about fourteen
inches long.

The makers of the instrument supply the most suitable oil, and
give the purchaser directions for occasionally oiling certain parts of
the meclmnism, A broad earner s-halr brush should be used for
dusting the cylinder after its surface has been traversed by the
sharp style of the recorder.

2. The wax cylinders are supplied each in a separate cardboard
box, which is lined with cotton-wool. Spoiled cylinders cannot
easily be used for taking new records. It is true that most phono-
graphs are fitted with a sharp cutting edge for the purpose of shaving
the surface of useless records, but so much practice is necessary
before a clean even surface can be obtained that the shaving
mechanism should, as far as possible, be avoided* The cylinders
are extremely fragile, but the manufacturers pack them so that the
loss due to breakage is negligibly small. In dry climates the

THE ETHNOLOGICAL STUDY OP MUSIC

261

cylinders keep well^ but in the damp a mould forms in and on
them, which seriously impairs the success of the record. I have
found that the cyUnders keep well in the damp heat of the tropics,
if each is wrapped in oiled paper and is enclosed in a tin case. Fresh
supplies of cyUnders, if wanted, can easily be sent out safely packed,

3, Recorders and reproducers vary both in type and in
efficiency. The purchaser should explain to the makers the kind of
music which he wishes to record, and should test the recorder before
employing it in the field*

4, It is important that the cylinder should revolve at the same
rate during reproduction as during the taking of the record ; for
the speed with which a phonographic record rotates determines not
only the tempo but the pitch of the sounds which it reproduces.
When a note of given pitch is sounded before the trumpet at the
time of taking the record, and when a note of precisely the same
pitch is later reproduced by tliat record, we can be assured that the
cylinder is rotating at the same speed during reproduction as it was
during the taking of the record. Accordingly, a pitch-pipe, such
as is sold at the music shops, should form part of the phonographic
equipment. This, when blown, emits a tone of definite pitch, e.g.
a - 435 vibrations per second. Just before any desired record is
taken, this pitch-pipe is sounded before the trumpet. Of course the
clockwork must not be stopped or its speed altered after the pitch-
pipe has sounded.

5, A spare recorder and reproducer should be taken, as well as
spare glass diaphragms and cement, in case of breakage. It is easy
to replace the broken glass disk of a reproducer or recorder and to
cement the style to it A spare trumpet, oil-can, brushes, and pitch-
pipes should be taken in case of possible loss. It is also advisable
to take certain screws and leather parts of the phonograph in
duplicate. A screwdriver should be included in the outfit.

Before a phonographic record is taken in the field, it is advisable
to hold a rehearsal of the performance, especially if the singers
or performers are inexperienced. Individual voices will be found
to differ considerably in the successfulness of the records which they
yield* A powerful voice wiU often yield a most unpleasantly sound-
ing phonogram. This is particularly apt to occur if the singer be not
placed so as to sing directly into the centre of the trumpet. If he
be sitting sideways near the instrument, so that his voice falls ob-
liquely on to the trumpet, a very jarring and unfaithful record will

252

THE ETHNOLOGICAL STUDY OF MUSIC

result. If a group of singers or an orchestra of instruments be
making the record, they should be grouped in a semicircle before
the phonograph, the most important soloists being placed nearest to
the mouth of the trumpet. When the piece is made up of several
simultaneous parts, each with a more or less independent tune or
rhythm, it may be advisable for the investigator to take more than
one record, placing now one singer or player {e.g. drummer), now
another in the foreground of the phonograph.

The speed with which the cylinder rotates varies with the extent
to which the clockwork is wound up, and can be regulated by means
of an adjustable head attached to the instrument. Before a record
is taken the clockwork should always be fully wound up. The rate
of rotation of the cylinder should roughly be two revolutions per
second. It can easily be gauged by lightly placing the finger upon
the small wheel over which the leather band passes. The rate
should be faster for music in which the tones are prolonged than for
music of a more lively and less monotonous kind, but the speed just
mentioned will be fomid generally serviceable.

The title of the song should be sung into the phonograph before
the record is taken. This is to be followed by the sounding of the
pitch'pipe, which should serve as a signal for the musicians to begin.
A number should be allotted to each record taken. The title and
the number of the record should be written on the outside of the tin or
cardboard case of the record, and should correspond with the number
in a note-book, in which are written the names, tribes, &c., of the execu-
tants, the instruments used, the significance, words, &c., of the music.

The phonograph should be similarly used to record the sounds
of instruments which cannot be relied on to keep their pitch when
they are sent to Europe. All stringed and reed instruments come
under this head, and such percussion instruments (e. g. gongs) as are
attuned by the attachment of pieces of wax. Even when the necks
of the stringed instruments are provided with frets, the performers
frequently vary the intonation by slightly changing the position of their
fingei-s. Similar variations are sometimes produced in the case of
wind instruments by only partially uncovering the holes.

Records when once taken should be reproduced as seldom as
possible. It is advisable that they be returned home so that the
records may be mechanically copied on to other cylmders without
needless delay. Or permanent moulds may be prepared from the
originals, and duphcate cylindei*s can be made from the moulds.

THE ETHNOLOGICAL STUDY OF MUSIC

253

When properly cared for, the wax records last for a very con-
siderable time without showing serious signs of deterioration.

TJie iranscription of phmiograms. — For the purposes of transcrib-
ing the phonographic records, two instruments are necessary. The
first is a metronome for determining the tempo of the music and
its variations. The second is some fonn of apparatus which will
produce tones of any desired pitch within a given range. Either an
Appun's ^Tonmesser' or a Stern's * Tonvariator ' can be employed
for this purj)ose. The former consists of a box of metal tongues, any
one of which can be made to vibrate at will by means of air driven
from bellows. The tongues need to be carefully tuned so as to give
tones successively differing by one or two vibrations. The latter is
a vertical cylindrical vessel provided with a narrow upper neck, over
the top of which a blast of air is driven, throwing the air contained
within the vessel into vibration. The pitch of the tone thus emitted
can be varied by diminishing or increasing the height of the
cylinder. The base of the cylinder consists of a movable plate, the
position of wliich can be delicately adjusted by a rod and screw
action attached.

The phonographic records should at first be roughly transcribed
in what appears to the observer to be the most nearly corresponding
notation. Then the pitch of the most important and prolonged tones
of the tune is carefully determined by comparison with the tones
of known pitch produced by the Tonmesser or Tonvariator, Any
given tone can be prolonged on the phonograph by holding up the
lever which plays upon the spiral steel tliread. By this means the
reproducer, instead of travelling along the spiral groove cut in the wax
cyUnder, remains stationary, continuing in the same groove and repro-
ducing an unchanged note while the cylinder is rotating. But such
procedure, if unduly prolonged, converts the spiral groove of the re-
cord into a circular groove, and so causes serious damage to the record.

Graphic recwrf^,— Graphic records of a tune may be produced
in the field by an arrangement which allows the vibrations of the
recording style to be written on a travelling sheet of smoked paper,
.Such smoked surfaces also afford valuable means of recording com-
plex rhythms. In place of the drum, stick, or rattle, a Morse key
is provided, and the taps made by the performer on this key are
electrically communicated to a * time-signal ' which is brought to bear
on the smoked surface. Below these markings another time-signal,
electrically connected with a silent clock, marks fifths of seconds.

THE SIGYNNAE OF HERODOTUS

AN ETHNOLOGICAL PROBLEM OF THE

EARLY IRON AGE

By X L, MYRES

The time has now gone by when it was safe to jeer at Herodotus
a mere retailer of travellers' stories. For us the Father of History
is no less the Father of Anthropology, That he is becsome so, for
some of us, is the outcome of a request made in all diffidence by
certain Oxford undergraduates, in the Easter Term of 1892, that the
Reader in Anthropology would lecture, if only for once, on the
earlier books of Herodotus, or at least on such passages of them
as demanded anthropological commentary; No one, I think, of
the audience of those lectures on * Anthropology as related to Ancient
and Modern History', has forgotten the wealth of learning, and
the truly planetary outlook with which that experiment was
made. It was obvious— as we had indeed suspected, but in far
greater degree — that there was a great mass of new material
already available for the interpretation of the ancient Mediterranean
world, and of a Greek's outlook over it, which would well repay
research ; and one at least of that audience has never regretted
the devotion of his life to that task,

The passage of Herodotus, which forms the subject of this
paper, must be studied in its context.* Herodotus has been
describing Thrace, of which the northern boundary is, for him, the
river Ister. ^ Further tioHh of this country no Ofie can say fiw certain

^ Hdt. V» 9 To ^c 7rpo5 pofy€*Mi rrj^ X'^P^^ *'''' Taimyr ovStU €j(tt iPpaaai ro arp^Ktt on-tvcf
ctcrt av$fnuwoL oikcovtcc avnjv, dAAa Ta "Jriprfv iJSiy rov "larpov fpT^fxos X^**P^ ^uiVrrat
ioikra Kat airctpos. fJMvvov^ Se ^vvapjiL iru^fcr^at olKiovra^ Tftprfv rov 'larpov avBptaTrovf:
TOtcrt ovyopLa €lvai %iyvvva%t €<r0^i Sc ;^ca>/icVoiis Mt^SiiqJ* tou? 8« 47r7rov9 avtiiiv ctfat
XatFiov^ aira*' to trihfjLa iwl wivrt SdicrvAov? to /3a.Bo^ rtiv rpi^utv, fiiKpovs: Bk icat frtfiov^ *cat
dSori/uTOif? dvBpa^ ^ip^iVt jciryw^cvovs Bi vtt appxira flvot o^vrdrov^" ap/winjXarcciv S< wpo^

tlvat Sc Mij8<ijv (T<fi€a^ dirotKOU? kiyoMTt. oKta^ S< oi^TOt MTifSwy dwotKot y€yoi^o"t, cy«tj fiiv
QVK t)^m i7rtif>pdu-a<r$ait ytvoiro S' ai' irav iv tw fxanpi^ )(p<iyia, (Ttyvwa^ 8* my KoXtowri
Atyves ol avtn hrkp Mao-craAiijs owtovre? tovs KamjAous, Kvirptoi Sc to. Bopara.

25S

THE SIGYNNAE OF HERODOTUS

who the human inhabiktnts are^ hut the parts Imyond the Ister seem
to be desert coutUry atid Hmitle^, TJie only human beings that I can
discover living beyond the Ister are those ivhose name is Sigynnae,
They imar Median costume. Their horses are shaggy all over^ to Jim
fingers^ depth of hair : they are sftiall and snub-nosed^ and not strong
enough to carry a man ; hut when they are harnessed to carts they are
very fast ; and that is why the natives go about in carts. This people's
frontiers reach nearly to the *Eneti on tJie Adriatic, They say^ limvever^
that they are a colony of Medes. How they have come to be a colony of
Medes I cannot explain : yet anything might happen^ if you gave it time
enough. I will only add timt the Ligurians limng tip country above
Marseilles call the pedlars ** sigynnae **, and the Cyprians their spears.

Here we have seven points which we must treat separately in
detail :— ^1) The Sigynnae live beyond the Danube and north of
Thrace ; (2) From the Lower Danube they extend westward as far
as the region of the 'Eneti on the Adriatic ; (3) they wear * Median
dress * ; (4) they beMeve themselves to be a colony of Medes ; (5) they
drive small, shaggy, snub-nosed ponies, very fast^ but too small for
riding ; (6) their name is applied to ' pedlars ' by the Ligurians
inland of Marseilles, and (7) by the people of Cyprus to some
pecuhar make of spear. Does all this information really belong
together ; or is Herodotus merely gaiTulous and incoherent ?

(1) The Sigynnae live beyond the Danube and north of Thrace.
This would seem to indicate the modem Wallachia or southern
Roumania, No other writer of pre- Alexandrine date mentions the
Sigynnae at alL But ApoUonius of Rhodes/ whose geographical
knowledge is extensive, gives %CyvupoL {Sigynni} as a people who in
early times lived near the island of Peuke — ^one of the numerous
eyots in the Roumanian section of the Danube. With them
ApoUonius associates Sptji^tp /ityaSc? tKvdai ; he regards them there-
fore as lying on the debatable land between the * Thracians ' of the
Balkan highlands and the * Scythians ' of the Roumanian and South
Russian steppes,

Strabo, the only other ancient writer who seems to mention
them, does not record them in this I'egion at all, but describes Siginni
in some detail in a Hst of the peoples of Caucasus, Nothing would be

^ ApclJ. Rhod. Argonaulkay iv. 320. I venture to assume that, for our present
purpose^ the variation lietween Jiiyvi^ai (Hdt), Siyivi'oi (Ap* Rh.), and ^ytyvm (Strabo)
iB negligible.

AN ETHNOLOGICAL PROBUEM

267

more unsafe than to base any inference on the mere recurrence
of a tribal name ; but Strabo is a learned and careful writer, and his
description of the Caucasian Siginni corresponds so closely with
the Herodotean account of their Danubian namesakes, that despite
the great interval of space and time, we seem to have something
more than a coincidence of sound, * In other respects the Siginni live
like Persians ; but* — unhke the Persians — ^they me ponies^ small and
shaggy^ which are not strong enough to carry a rider ; and so tliey drive
them in Imniess^ four-ifirhand' ^

Even this of course might be interpreted as a blundered
plagiarism from Herodotus, introduced to * add verisimilitude ' to the
bald and unconvincing fact of the duplicated name- But this
explanation is precluded by the circumstance that these Caucasian
Siginni live adjacent to a group of peoples whom Strabo describes
as taking exactly the same pessimistic view of life as the Thracian
Trausi of Herodotus : yet in this instance the synonym-motive is
absent ; and under these circumstances it is not easy to see why
Strabo should be suspected of having copied in the one instance any
more than in the other, especially as in neither case does he seem to
be aware of the parallel'^

(2) From the Lower Danube the Stgynnae estend westtvard as far as
the region of the 'Eneti on the Adriatic, We have seen already that for
Herodotus the proper home of the Sigynnae is in Wallachia ; but he
describes the country beyond the Danube as * lioiitless '—that is,
devoid of natiural barriers — and regards the Sigynnae as extending,
with their ponyKiarts, nearly to the head of the Adriatic : for the

* Strabo, 520 ^lyiwoi 8e r^EAAa pxv wc/ocrtfotwrcy, Imrap/oi^ 3c )(pSiynu fAucpoi^ Saxria-iy
^n€fi tinrcmjr 6)(tlv filv ov E/vvarat, r<^p«r7ra» Si ^tvyyvova'ty>

' Herodotus, v. 4 Tpavcrol Sk to, fikv aXXa tcara ravra roUrt 5AAo«7t ^p^t^t hnrikioMTh
tiara Bi tov ytvofitvov cr<^t kuX awKiytvofiivov Trotcwi TOioSc* tok /tcv ycvo/to^i' 7r€ptti6pL€voi oi
irpwTiqKovT€^ 6Ko<fyvpovTai^ otra fitv &i lfr€ir€ iytvtro Avawk^nat hoko, dn^co/icvoi ra
ar^punr^ta rnvra waBia* rov S* iliroy€v6fji€vov TratfovrcV re teal yjMfjitvot y^ Kprnrrovint
cTTiAcyorrcs wriitv t^aKtiiv iiairaXXa^Bil^ iarl iv ttoot^ fv^atfjuovij}* Compare what he says
of the Getae, who lie, Hke the Sigynnae, across the Danube opposite to Thrace, and
* profess to live for ever ' (Hdt. iv. 95).

Strabo's contribution is, that he applies to certain * quite barbarous' peoples of
' Caucasus and the rest of the highland countty * the well-known lines of Euripides
{Krcsphontcs, fr. 452 Nauck) : —

TOK ^vvra $pnfp^€tv cIs 5cr' tp^tTai KatttL,
riv y av ^avovra kol irvvtav wrrravp^lvov
)(aipo>VTai wvtftyffjLOvvTtK itnr€p.W€iv ^pnav*

268

THE SIGYNNAE OF HERODOTUS

word KaTTjK€iy in Herodotus lias regulai'ly the sense of ' extending
seawards \ It is safe therefore to infer that in spite of his allusion
elsewhere to *the rivers Alpis and Karpis' as tributaries of the
Danube,' he had no clear knowledge of the existence of the Kar-
pathians or of the Julian Alps, and regarded the steppes of South
Russia and Roumanla as continuous with the Hungarian plain^ and
probably this plain also as continuous with the flat land round Venice.
And so far as the actual distribution of nomad or semi-nomad
peoples is concerned^ he was indeed more nearly right than he knew ;
for neither the Karpathians nor the Julian Alps have ever interposed
a serious obstacle in face of such peoples.

There was excellent reason why Herodotus should apply the
qualification * on tim Adriatic ' to the name of the Eneti ; for it was
necessary, even in his time, to distinguish these Adriatic 'Eneti, or
as later geographers call them, Veneti^ in the flat land between the
Timavo and the Adige^ from those 'Eneti whom Homer describes as
settled in Paphlagonia in North- West Asia Minor;* an extreme
instance of a large class of duplicated tribe names to east and
to west of the head of the Aegean Sea« Later geographers, after
Caesars time,^ had also to take into account the seafaring Veneti of
Brittany, but of these* as we should expect, Herodotus has no know-
ledge.

Among these later geographers there was general agreement
that these scattered bodies of Veneti had some original unity ; and
in Strabo's time theories were current. Either the Adriatic VEneti
were a Cisalpine ofehoot of the Breton Veneti,^ or they were colonists
(under the Trojan Antenor) of their Paphlagonian namesakes,^ and
a branch of that great Trojan or Phrygian exodus, the best known
legends of which underlie Virgil's story of Aeneas. But all agreed
that these Adriatic VEneti were intrusive in Italy ; and this gives
point to tho phrase of Herodotus that the Sigynnae likewise * extend
seawards ' on their track, * as far as the VEneti on the Adriatic/
For it almost looks as if he had in his mind the third possible, view
of the relationship of these two groups of Veneti ; namely that both
represent fragments of an earlier continuous Venetia, which had

' Hdt iv. 49.

* Homer, Iliad ii 852.

* Caesar, B. G, iL U,

* Strabo, 195, 216 ; Ptokmy, ii. 8. 6 ; Dio Casaius, 39, 40.

* Strabo, 61, 195, 212, 543, 008 ; Livy, L L

AN ETHNOLOGICAL PROBLEM

269

been shattered and pushed right aad left by the advent of the
Danubian Sigynnae.

I think it will be clear from what precedes, that any attempt
to explain the Sigynnae of Herodotus must take account of a
geograpiiical distribution which extends, north of the Danube, from
a point near the head of the Adriatic to the neighbourhood of the
Danube mouth ; and includes also a district in Caucasus where there
survived not merely the name Siginm but also customs and beliefe
which prevailed^ earUer, along the Danube. The Paphlagoniau
VEneti, in their Thraco- Phrygian context, supply probably a clue as
to the way by which Strabo's Siginni i-eached their Caucasian home.

(3) The Sigynnue wear Median dress. This costume, which for
Greek writers means the costume of a Persian in the sixth and
fifth centuries b.c., is described by Herodotus in some detail.^ Its
principal characteristics are the Kvp0ao-iiq or rtapa^ a soft conical cap ;
the /caumiaj?, a sleeved jacket or coat, described also by Herodotus
as a KiBiiiP ()(tTQij/) ^tiptZorros ; and the ava^vplht^;^ or trousei'S. It is
the latter garment which, being most foreign to Gi^ek custom, caused
the greatest remark, and may be taken to be mainly intended here.
For the same word ava^vplht^ — perhaps Persian, and certainly not
Greek -^-is used regularly in later Greek to describe any kind of
trousers, whether worn by Orientals or not. Polybius, for example,^
uses it three centuries later than Herodotus to describe the nether
garments of the Insubrian Gauls of the neighbourhood of Milan,
which Latin writers— equally unprovided with a word of their own
for such clothing — denote by their Gaulish name of braccae or
* breeches *. These hraccm were recognized, in due course, as being
characteristic of Gauls in general ; and we find Gallia Braccata used,
as the popular correlative of Gallia Togata^ to denote those parts of
Gaul which had not yet adopted Roman civil costume.

Strabo's Caucasian Siginm are not expressly equipped with
dwi^/otS€9 or braccae; but when he says* that except for their pony-
carts *they miitate the Persians', it is natural to suppose that this
conspicuously * Persian ' trait was not wanting. That they ' imitated
the Persians' it would be superfluous to mention^ if this did not
distinguish these Siginni from some at least of their Caucasian neigh-
bours. We are left, therefore, with the probabihty that these Cau-

» Hdt. I 71, 185 ; iii 84 ; v. 49 j vil 61.

* See Baehr^B note on Hdt i. 71.

* Strabo, 520 rlKXa fxtv Trcpo-tfovo-c,

s a

Compare Xenophon, Anabasis, L 2. 27*
* Polybius, ii. 136. 7.

THE 8IGYNHAE OF HEEODOTOS

Si^nm formed a more or Imb iBolated pafccli of people who
1 PeniMW in eome iBspecte in which the Danubian Sigrnnae
nmuM ed Aem; but at the same time resembled the Danubian
► in another aspect — namely tbetr pony-carts, — in which
pwrFn^^ was wanting. As between Persian and Danubian
aUrntm for the ' PeiBian ' attributes of the Caucasian Siginni, the
balance of probability is on tlie whole in fiivour of the Danubian
connexion*

(i) The Sigjfmiae bdiem tkemadcm ta be a cohng cfMedes. This
is, of course, very far from an aBsertioii thai they reaUy were so ; it
is dilBeult to see under what cirenmstaiiees such a colonization could
have taken place; and Strabos account of Siginni in Caucasus
suggests that any movement which has occurred has been in the
other direction^ from South-Eastem Europe into Asia Minor. There
WM| moreover, later at all events, a notable people called the Maedi
(MuiSoi ') in Western Thrace, who may have been kinsmen of the
Sigynnae, and may well have been confused, in local speech, with the
Medi (M17S04) whose name would suggest itself at once to Greek
observers, to account for the Sigynnian trousers,

Herodotus is inclined to admit a Median origin, only if time
will allow it But this is precisely what time will not allow. The
Modes themselves only appear in Media — if we may judge from
Assyrian records of them^ — a little before 700 rc, ; and there is no
reason for believing that the Median Empire ever extended further
west than the Halys, which became its frontier, after a six years'
war with Lydia, under the treaty of 686 a a There is on the other
hand some reason, I tliink, for believing that one element in the
turmoil of nationalities, which is the background of the last century
of Assyria, was an eastward stream of migration through Asia Minor
into Armenia and probably even further to the south-eastward;
and that this eastward movement out of Asia Minor is itself to bo
identified as the sequel of an eastward movement into Asia Minor by
way of the Hellespont, of which one of the latest incidents is the

* The Maedi are first recorded by Thucydides (ii* 98) at the end of the fifth
MCtury B* c. He places them west of Paeonia (the upper valley of the Axius) aad
near the Sinti of the middle Strymoo* Later writers find them also further north
round the headwaters of the Axius and the Margus. Polyb* x. 41. 4 ; Diod. Sic.
XXX, 19; Liv. xxv, 25, xxvi. 26, xl. 22; Plin. iv. 40; Ptol ill. 11, 8; Eutrop.
V. 7 ; Justin, 15, 2, and Diodorus, loc. at, actually write Medi and Mrj^un].

AN ETHNOLOGICAL PROBLEM

261

Kimmerian invasion of the seventh century, and one of the earher
was the Thraco-Phrygian movements of approximately Homeric time/
of which we have ah-eady to take note in dealing with the Paphla-
gonian VEneti.

(5) The Sigyyinae drive stnaU, shaggy, snuh^nosed ponies, very fast,
hit too small for riditig. It is a graphic description of the typical
breed of northern Europe. With the snub^nose compare the 'great
and hooked head * of the Hungarian horse of tiie fourth century a. d./
coupled as it is with low stature, copious mane and tail, and great
speed. There was also as late as Roman Imperial times a notable
breed of horses of great speed and endurance among the Adriatic
Veueti.^ Similar dwarf horses (ginni) were a regular article of export
among the Ligurians of Savoy/ and on the testimony of Aristotle ^
the breed can be traced in this region as far back as the fourth century.^

At this point it is remarkable that, as we have seen, Strabo s
Caucasian Siginni likemse * use httle shaggy ponies which are unable
to carry a rider ; so they harness them four-in-hand '.

(6) The name Stgyntme is applied to pedlars by the Ligurians * inland
of Marseilles \ i.e. up the Rhone and its eastern tributaries Isere and
Druence. Now immediately beyond this * Ligurian * country cornea
that of the GalUc AMobroges, and beyond these and the Rhone he the
Jura and the country of the Sequani ; and there is clearly sufficient

'superficial similarity between the names Heqtmni and Xiyvvpai to
justify further inquiry.

On the purely philological side we should note that the word
a-tyvvvo<; when used for a javelin, as Herodotus describes, though
common in later Greek,** remains very variable in its spelling,^ and

' Homer, Hiad ii. 845 ; uL 184 ff. See alao a paper entitled A Hisiori/ of the
PeJm^ian TJieory in the forthcoming volume of the Journal of Hellenic Studies,

* Vegetiiis, Ars Vet. iv. 6. 5, quoted in full by Riilgeway, Origin and Influence
of the Thoroughbred Horse^ pp. 318-19, Compare also the ' ugly shaped skull * of horse

I skeletons from Macon, quoted by Ridge way, he, ciL^ p, 94, On the Sigynnae see also

^^ Eidgeway, loc* cit, p, 345,

^fe • Strabo, 215. * Strabo, 202 ; Ridgeway, loc. cit,, p. 321.

^^^^L ^ Aristotle, HisL Animaliumf vi. 24. 1 ; de Gen, An. ii, 8. 24.

^^^^F * For its use in the Cypriote dialect and for the kind of spear which it denoted,

I Bee below p. 271,

I ^ The regular variants are <riyvvvo^ {mymTa^ u-iyvwov), iriyvfivov^ friyvvTf {(riyvvrjs;)i

^^ Q-ipvvri^ iifivviq. See stein a note on Hdt v, 9. It is perhaps worth noting that

■

2«2

THE SIGYNNAE OF HERODOTUS

this in turn suggests (a) that it was recognized as a foreign word,
not easy to pronounce^ and (h) that the difficulty arose from the fact
that the y represented, not a pure G sound, but something between
this and a labial (represented by yfi or even fi) ; in fact, what we
denote by a Q,

On the archaeological side^ the /camjXot of the Ligiirians, whether
Sequani or not, must have had something of value to export, either
from their own country or from beyond it. Now the Sequani do not
lie directly on any of the great northern routes from Marseilles, but
rather in a less passable highland region between these routes.
But they do lie in a region which produced in classical times at
any rat^ one valuable commodity, the iron of the Jura forges so
fully explored by M. Quiquerez.» Now these forges of the Jura lie in
BO intimate a coimexion both of place and date with the copious
, iron-using culture of La Tene and similar sites in Western Switzer-
land, and these in their turn with an important group of early
Carinthian and Styrian iron-workings,- and even with a site as far
east as that at Gyalar in Transylvania, "^ that we are tempted to
pursue this hint and see whither it will lead us. Was there, in fact,
any reason for describing any particular kind of iron javelin by the
name of the Sequani or their country ? *

La Tene J tJie Sequani^ and the type of the * Gaemim\ The La Tene
phase of iron-using culture does not itself go back to the beginning of
the Iron Age in Europe, It is a late and rather local development

in a fragment of the late comedian Baton (AthenaeuB^ xiv. 662 c) occurs a woman
named ^1^^107, apparently a cook, and probably a slave. Was she perhaps by birth
a Sigpnna^ or even a Seqimna ?

' Quiquerez, * Notice sur les forges primitives dans le Jura/ in MUth, d, antiq,
GeselMm/t von ZUrich, 187L

* Qowland, Joum. Iron and Steel Institute (London, 1897), liL 206, On the
whole subject see also Beck, G^schkMe des Eisens (Braunschweig, 1892), and Gk)w]and,
' The Early Metallurgy of Copper, Tin, and Iron in Europe/ in Ardtaedogia, Ivi
(1890), pp. 315-21.

' MUnichadtirfer, Der EUtttnherger Enberg; Wankel, PrUfHstorische Eisen-
schmelZ' und SehmiedestCdten (Wien), 1879.

^ It was only while correcting the proof of this essay that I became acquainted
with the article of M. C, jullian (Mevue des i^tudes AncienneSf viii (1906), pp. Ill E),
in which he discusses the ethnology of the Trans- Alpine iron-culture, and attributes
the culture of Hallstatt to the Sigynnae. On the Sigynnae as Banubian traders —
* seraient-ils lea premiers importateure du bronze en Occident?'-— see Bertrand and
Reinach* La Gauk avant ks Gauhis* Paris, lb91, pp, 259-60.

AN ETHNOLOGICAL PKOBLEM

263

of that earlier style which takes its name from the necropolis of
Hallstatt, Continuity of development is more marked in some
departments than in others, and is particularly clear in the case of
a large series of spear-heads, which show continuous development
from simple Late Bronze Age prototypes into two main groups of
forms.

The Bronze Age prototypes show a double-edged blade reinforced
along its median line by a kind of mid-rib, to resist lateral stresses.
At the base of the blade the mid-rib is prolonged into a tubidar
socket, into which the pointed end of the spear^haft is thrust, and
then made secure by one or more nails driven transversely through
holes in the wall of the socket. Sometimes the blade with its mid-
rib has a concave section like that of a hollow-ground razor ; some-
times the mid-rib is treated apart from the lateral wings as
a prolongation of the tubular socket, and has a convex section for
a considerable part of its length ; occasionally it is even hollow for
some distance beyond the base of the wings.

Even in the Bronze Age the proportions of socket, mid-rib,
and wings vary greatly ; the extremes of the series being furnished
(a) by a long and very narrow blade, with mid-rib enlarged and wings
reduced, and (i) by a short broad blade, with leaf-shaped outhne, and
wings greatly expanded. And these opposite tendencies are still
active at the beginning of the Age of Iron.

The short broad series leads of course to an almost infinite
variety of forms. It is well represented at Hallstatt, and in most
other sites of that phase j and at La T6ne it leads to a remarkable
school, in which the wings become unsymmetrical and are worked
into flamboyant and fantastic outlines or thinned away into large
apertures internally. This series, however, remains rather local in
Western Switzerland ; and it betrays an ability to cater for
individual tastes, which could hai-dly have existed except in close
proximity to the place of production ; in this case, to the iron
workings of the Jura,

The other series begins, at Hallstatt, with a weapon of which
a good example is figured by von SackenJ The blade is about
18 inches long, and nowhere more than Ij inches wide; and the
cutting edges are parallel for nearly the whole of their length.
The wings are thus veiy shallow in themselves ; and their shallow-
ness is accentuated by the extreme prominence of the mid-rib,
' von Sackeo, GraberfcU mn Hallstatt (Wien, 1868), PI VII. 2.

THE SIGYNNAE OF HERODOTUS

wiiieli is namofr and deep, and tapers to its cn^ almost as sharply
m do the wings. The crosd-aection of the blade is in fact almost
eruciform, and the general effect is that of a four-winged blade, ^ In
eompoctneHy penetratioQ, and focility of withdrawal from the
wound, this type compares not unfaTourably with the French
duelling-sword, and with the long bayonets of the nineteenth
century. As it has reduced its external excrescences to a minimum,
it can be readily packed in large bundles, or held in a cluster in
the hand, without inconvenience or waste of space : it is therefore
eminently portable, and adapted to serve as missile ammunition ;
while its form recommends it equally for throwing or for spear-
play. It is not surprising therefore to find that this weapon has
a wide distribution over the Hallstatt region ; ^ that it recurs over
a considerable range of time, in Switzerland ; ^ that, where it occurs
in tombs, it is frequently found two or more together, after the
manner of a throwing-spear ; * and that this equipment of hght
throwing-spears was recognized by the men south of the Alps as
characteristic of intruders from the mountains themselves or from
beyond."

As the absolute chronology of the Hallstatt culture is still
matter of dispute, and as it strengthens our aigument considerably
if this long narrow-bladed type can be shown to have been per-
manently established in this trans-Alpine area, it is well to note
here that another practically mngless spear was evolved also in
the La Tene cultm^e, in which one of the best defined varieties of
spear-head has its long tubular socket (or rather the neck which
joins it with the midrib and wings) enormously developed at the
expense of the rest of the head ; so that the wings extend at most
over half of its total length, and occasionally over less than one
quarter. In this type, however, although the wings are so short,

' Curiously enough, this type was anticipated in Cyprus quite early in the
Bronze Age (Ohnefalsch-Richter, Kypros^ the Bible and Homer (London, 1893),
PI, oli, 27) ; but like so many things in early Cyprus, it does not seem to have
persisted.

' von Sacken, loc. ciU, p. 36,

• Troyon, Habitations Locudres^ PI. xv, 1, 2, 5, 6, 15, 17 ; Desor, P/ahlbauten
ies Neuenburger Sees, p. 1^8.

* Instances in von Sacken^ loc. cit, p. S6.

^ Caesar only uses the word gaesum once, in recounting a fight in YalatB
against strictly Alpine folk, B. G. iiu 4. 1. Compare Virgirs phrase, Duo qui$qm
Alpitm mruscani ijacm numu.—Aeneid viiL 661.

AN ETHNOLOGICAL PROBLEM

265

and proportionately narrow, they are still regular wings, with a
lanceolate outline, and the well-marked concave surfaces which are
8o characteristic of the speai-s of La Tene. The wingless neck, too,
in this series remains cylindrical, and never becomes cruciform or
foursquare.^

Not only was a Ught throwing-spear of these or similar fabrics
recognized in Italy as characteristic of trans- Alpine warfare : even
the name of this trans- Alpine weapon, gaesumj yala-ov^ passed early
into Latin, and thence into the Greek of Polybius ; and it is not
improbable that the GaesataOj a great free-company from the tians-
Alpine country and in paiiicular from the valley of the Rhone, ^
who caused dire terror in Northern Italy in the tumultuary years
between 226 and 222 b.c., owed their own name like^vise to their
characteristic weapon.^

The identification of the gmsum vrith this particular type of
spear is due to Hesychius, who explains it as ifx^oXmv oXocrtSTjpoj/,
*a spit-like weapon, wholly of iron; The Hallstatt spears are of
course not * wholly of iron' in the strictest sense, for they are
socketed to receive a wooden shaft.* But a spear with so long and
narrow a head is obviously in far less need of a long shaft to steady
its flight ; or rather, any shaft long enough to do so would be too

' For examples see Gross, La I^ns (Paris, 1887), PL V. 8. 7 (one-half length) ;
v. 1. 11 (one-quarter); reproduced in Munro, Lake DwelUngs of Europe (London,
1890), fig. 88 (19, 20, 21),

' Adjacent, that is, to the iron-working Jura and the later home of the
SequanL

' Polybius^ it is true, says (ii. 22. 1) that Gaesatae meant * people paid to fight ',
i. e. specifically mercenaries ; but nothing is so common as the designation of such
troops by their equipment, particularly as this so frequently stands in contrast with
that of the people whom they happen to bo serving, O rosins (4. l$\ on the other
hand, calls them gaesati^ and clearly regarded the name merely as an adjective
formed from gaesum. The word gaesum, formerly connected by Schrader with
Gothic gasas, * spike' (Schrader-Jevons, Prehistoric Antiquities of the Aryan Peoples,
London, 1890, p. 218} is now derived by him from a practically identical Gallic
word, which reappears in the Irish gae^ * spear/ and has pi-obably been borrowed
by Teutonic speech (OHO, ger^ AS* gdr, &c.). Eeallexicon der Indo-Oermamschm
Altertumskunde, Strassburg, 1901, p. 787.

* Actual examples of spears * wholly of irun '—shaft and blade in one piece —
occur sporadically all through the iron- working culture of Equatorial Africa which
presents so many similarities of technique with that of Early Europe ; thei-e are
specimens in the Pitt^Bivers Museum and in the Leicester Museum. I do not,
however, know of any European instance.

266

THE SIGYNNAE OF HERODOTUS

long to be portable or convenient either for throwing or for
thrusting.

The ^ $igynna\ the ^ gaesum\ and the Moman *pilum\ This last
point will become clearer when we trace the distribution of the
narrow-bladed type on the margins of the Hallstatt region -

The crucial instance is in Spain, Athenaeus (vi. 273) states
that the gaesuni^ which in Polybius s time was still a regular throwing-
spear of the Roman legionary, was borrowed by the Romans from
the Iberians of Spain. Fas est et ah haste docerL It was not the
first time tliat the Romans had remodeUed their mode of warfare
in imitation of the practice of an enemy whom they respected.

Moreover the piUim itself, the regulation throwing-spear of
the Roman legionary, is nothing but an extreme case of specializa-
tion in a derivative of the narrow-bladed throwing-spean To
demonstrate this point in detail is difficult, for no extant pilum
from a Roman site is certainly earUer than the Numantine War of
148-133 B.C,,' and the description given by Polybius belongs, so
far as eyewitness goes, to the same generation of men, and at
earliest to the generation after the Roman occupation of Spain.
Consequently we have no evidence as to the antiquity of the earhest
form of pihmi which is directly known to us.'^

Most modem archaeologists still follow Lindensclmiit's opinion^
that the pilum came to the Romans from Etruria This opinion —
based on one specimen of uncertain date, from Vulci,^ and supported
subsequently by one, better authenticated, from an eai-ly sixth-century
tomb in the Museum at Perugia — obviously takes the problem only
one stage further back, and leaves unsolved the question how or
whence the Etruscans acquired the pilum. A comparison, however,
of these rare early pila with the later varieties, and also with the
trans*Alpine types to which we have seen reason to assign the name
gaesum^ makes it possible to reconstruct the main outlines of its
development.

* These Numantin©jji7a are referred to in Jahrh. Kais, Arcluieol Jmtituts^ 1907,
Anseiger, p. 84, but are not yet published fuDy.

" The cliisaicAl descriptions are in Polybiua, vl, 23, 9-11 ; DioDj^ius of Hali-
cam&ssus, Aniiq. Mom. 5. 4G ; Appi&n, CdtioOf 1 ; Vegeiius^ 4. 20, 2* IB,
The modern literature is collected in Mommfleo-lf arquand, Manud des Antiquum
Momainrs, xi (Paris, 1891), pp. 28-9.

* Die AUerthUmer unserer heidnischen Vorgeii (Mainz, 1864^81), iii, 6. 7. L

* Linden Bchnutf loc. cit ; Museo Grtgoriano^ Ixxjciv. 6.

AN ETHNOLOGICAL PROBLEM

267

The Roman pilum seems to have originated as an ordinary
throwing-spear, with a socketed head of the long naiTOW tjrpe.
Whether its direct ancestor is the four-winged Hallstatt type, or an
early version of the long-necked but stUl lanceolate type which is
perpetuated at La Tene, is a minor point : the four-square section
of the neck points to the former; the distinct though diminutive
head, so commonly prefixed in extant examples, suggests con-
tamination, at least, with the latter.

But before the time of Polybius the pilum had undergone two
principal reformations. In the first place, its head was very greatly
elongated, so that the winged point and the conical socket came to
be mere terminal appendages of an exaggerated midrib or neck, of
cylindrical or four-square section. This phase is represented by
the well-known specimen from Vulci, which though it is of uncertain
date, and by no means certainly a Roman pilum in the strict sense
at all, is of great value as proving the existence of this type in
middle Italy at all.

The second modification of the pilum is more important. A
long shaft, added to so long a head, would have been cumbrous;
yet some weight waa necessary to give the weapon a true balance.
So the shaft was made short and thick ; and this incidentally gave
a better grip to the legionary and increased both accuracy and
penetration. Tiie new shaft tapered sharply where it fitted to the
head ; and the socket was made wider and shallower accordingly.
But while the leverage of the long head and the weight of the shaft
increased, the reduction of the socket made this mode of hafting
quite inadequate, and led to the second reform, now to be mentioned.
Without discarding the socket — ^for a Roman never willingly dis-
carded anytliing which had once served any purpose — the head
acquired a shaftward prolongation in the shape of a broad flat tang
of varying length. This was inserted in a deep longitudinal cleft
in the thick shaft, and driven home till the conical socket closed
over the cleft point of the wood, and gripped it closely together.
Then the whole was pinned firmly by two or more iron nails
driven right through shaft and tang, or clamped externally by
metal rings or ferules.^

This is the pilum described by Polybius in the second cen-
tury B. c. Later improvements modified the temper of the point,

* The word Aa/?t? used by Polybius, vi, 23. 11, denotes any kind of *grip' or
" clamp *, and is applicable either to a ferule or to a rivet or nail.

268

THE SIGYNNAE OF HERODOTUS

the inordinate security of the haftiug, and the diameter of the
shaft, till in the second century a.d. the shaft was but little thicker
than the head. But the wide conical socket persisted long after
the shaft had shrunk, useless but significant, to cause interminable
trouble to the antiquaries.

There is, however^ some little evidence for the beUef that the
earlier Roman armies made regular use of a weapon which a later
Roman would have described not as a pUum but as a gaesum. Livy,
for example, in describing (rather obscurely) certain changes of
equipment in the generation which followed the great Gaulish raid
of 390 B,c,, says^ that each maniple of the legion included twenty
leves^ • . . qui hustani tantum gaemque gererent^ i.e. carried the hoplite-
spear, of Greek origin and contemporary fashion, and also two or
more light tlirowing-spears. The use of the pihim at this period was
apparently confined to the triarii^ for all the remainder were classed
together as antepilani; even this attribution of the piJum to the
triarii as early as the Latin War of 337 a a involves difiiculties,
unless the statement of Sallust that the Romans acquired their arma
atqtse tela militnria a Samnitibus- refers exclusively to the First
Samnite War of 843 a c, : the Great War, of course, did not break
out till 326 ac. By Polybius's own time, of course, the triarii had
ceased to use the pilum and had reverted for a while to the hasta.^

Another argument for the eai'lier use of the gaesum by the
legions comes from Polybius s statement that for killing an enemy
in battle a legionary was solemnly presented with a yauroi/.* If the
legions had always used the pilum^ it is difficult to see when or how
this custom could have arisen. If, however, the legions had at one
time been armed with an undifferentiated throwing-spear, borrowed

* Livy, viiL 8. 6.

* Sallust, CatRine, 51. 38. AthenaeuSi xL 273, while attributing (as we have
seen) the yaurov to the Spanish Iberians, confirms Sallust s statement so far as the
ahleld (Bvpto^yis concerned. Livy (ii. 30), on the other hand, describes the use (or
rather the exceptional disuse) of the pilum in a battle wluch he dates as early as
492 B.C. : but com}>are Plutarch s use of iVo-os in a passage {CamUlus, 40) where
Ihe context shows that the hasta is intended.

^ Polybius, vi. 23. 16. It was probably Marius who finally assigned the
filum to all ranks alike^ as we find it in Caesar's time.

* Polybiufl, vi, 39. 3. First an iyKt^fitor is pronounced ; /Acra S« ravra rm ^tr

T^Kiitrai^i TTokifuav yaurov SajpciTai^ T(J ^i icaTa^oAoi^Tt teal tncuXrwraKri, rw fitv ^ef *i» <fnaXrfv,

T<j» S* Jinrct (^KtAapa, €$ tip;^« 5c yato-Qv fiovov* The inference is that at on© time the
yawroy had been the regular weapon of all branches of the service.

AN ETHNOLOGICAL PROBLEM

269

from their fourth-century enemies the Gauls, it is easy to see how
the old-fashioned weapon^which remained in use throughout
among the auxiliary troops — should have been retained for the
ceremony; and how a phrase 'to win a gaesiim' should have sui'vived,
like our own phrase about * winning one's spurs \ long after the
prize itself had lapsed or changed its character.

But the best argument for the derivation of the pihwi from the
Alpine and trans- Alpine long-bladed gaesum is morphological. The
later pila^ of the first and late second century b.c., varied a good
deal in form, and in Polybius s time there were clearly two well-
marked types, one stouter than the other. Tlie former was evidently
the more specialized and remarkable, with its long needle-like point,
and its short, log-like shaft three inches or more in diameter.^ The
latter, in comparison, is undifferentiated and almost primitive, and
Polybius himself compares it with a * tri^vviov of moderate size':^ it
was carried in addition to the stouter type, ^ not as a substitute for it.
Like the stouter form it had a short shaft, and a very long narrow
point, of equal length with the shaft. The head, moreover, besides
being socketed, penetrated the shaft, as we have seen, with a long
flat tang for half its length. Of the completed weapon, therefore,
two-thirds were wholly or partially of iron, and only one-third was
wholly of wood. At the point where the iron joined the wood,
the diameter of the iron was about a finger's-width and a half. The
point, both of the stout and of the slender pilum^ varied in details ;
but all forms agree in consisting of (1) the long flat tang enclosed
within the shaft and riveted to it ; (2) the conical socket already
mentioned, enveloping and clamping together the cleft tip of the
shaft ; (3) an enormously elongated midrib, quite without wings, but
often square in section. This last feature was partly, no doubt, for
simplicity of manufacture ; but it acquires further importance when
we remember that the extremer forms of the long-bladed type in
the Hallstatt area rapidly approach a four-square section in proportion

^ This is the liteml sense of the words of Polybius* No known pilum, how-
ever, shows any such shaft, and it has been suggested by Lindenschmit {loc.ciLf iiL
6, 7» 14 a) that this dimension is that of the peculiar conical rim of the socket of
the pUum as shown on the tombstone of C. Valerius at Wiesbaden (Lindenschmit,
loccitf iii, 6, 5, 1). But the Wiesbaden monument belongs, at earliest, to a period
more than two centuries after the death of Polybius.

* Polybius, vi. 23. 9 ot yt /x^r Xryrrot <rt^iWot5 toiKatri trv^^Tpoi^,

• Like the gaesa of the fourth-century legionary in Livy, viiL 8. 6, v. aboTe,

270

THE SIGYNNAE OF HERODOTUS

m their wings are reduced, and that it would not need much further
specialization m the same direction to reduce both them and the
ridgee of the midrib to the appearance of mere ribs or angles on the
sur£ae6 of a rounded spike. Sometimes this spike was armed, in the
piiuMj with a diminutive head with wings or barbs, like those of
the long-necked spears of La Tene; sometimes the four-square
section ran out to the very point or was repeated on the head
above-mentioned ; sometimes the ridges or angles faded away into
a smooth conical point.

It will be obvious from this description how closely the fully-
developed pUum corresponded with the Hesychian description of the
gaesum as an cfx^oktov okocrlBrjpop^ * a spear like a spit, made wholly
of iron ' ; it had indeed a shaft, but for half its length this shaft
was reinforced with iron ; and it would be no exaggeration to say
that the weapon was * practically all point \ Moreover, the absence
of wings and the frequent absence even of the little barbed head
gave it a truly spit-like appearance.

The story that the Roman gaesum was borrowed from the
Iberians of Spain comes to us, as we have seen, from a source which
though learned, is late : ^ and it may i-easonably be asked whether the
Gallic tumuUus of 225 b.c, or even the great raid of 890 b,c., does not
supply a vermr causa for its introduction. In the absence, moreover,
of adequate evidence from Spain itself as to the spears of the Celti-
berian iron age, we are not in a position to prove dii'ectly that any
such weapon was ever in use there. At the same time, our know-
ledge of the ethnology of the peninsula would lead us to conclude
firsts that Spain suffered, very much in the same way as Italy, from
the GaUic movements of the fifth and fourth century b.c. ; secondly^
that tliis meant the incursion into Spain of a mihtant element in a
phase of culture which was transitional from the Hallstatt type to
that of La Tene ; thirdly^ that, onco established, this iron-using
culture made use of the copious iron-supplies of the peninsula to
release itself from dependence on the mines and forges of its place
of origin, and to develop the weapons which it brought with it in
original and more special directions, particularly if these at the same
time simplified the process of manufacture by the omission of such
refinements as grooves and re-entrant angles between ridges. These

' AthenaeuSr vi« 278. It may, moreover, fairly be questioned whether
Athenaeus ia not transferring to the pilum the well-authenticated derivation of the
legionary ^ladim.

EHi-.

AN ETHNOLOGICAL PROBLEM

271

are, in fact, just the conditions under which so special a form as the
Roman pilum may reasonably be thought to have been perfected ;
and they supply some indirect reason for accepting as probable the
ancient tradition that some form of Roman throwing-spear was of
Spanish origin, and was adopted by the Romans as the result of their
experience of Spanish warfare.

The considerations above stated as to the Spanish Iron Age
stand in obvious relation to the statement of Herodotus that the
Ligurians above Marseilles call their pedlars Sigynnae. Whither
were these pedlars going, and what were they going to sell ? The
very small evidence which we possess goes to sliow that Spain
entered late upon its Iron Age, and in particular that it depended
long, both for bronze and for iron, on external sources of supply. In
the North, at all events, the peninsula long remained under the
influence of the culture of the continent to the north-east. The
statement of Herodotus, therefore, which belongs, of course, to the
century before the great Gallic inroads, is of the greatest value as an
indication of what actually was going on : and the gist of it is this,
that the Ligurians above Marseilles gave to the men who worked the
transport-trade across their country a name which for Herodotus is
that of a Danubian people. This transport-trade from the Danubian
region into the Rlione basin was clearly in a westerly direction ; and
out to the west, for Herodotus and his contemporaries, lie only the
Kelts, the Kynesii (whom Herodotus makes the most westerly of all)
and the Iberian population of Spam, Here again we can prove
nothing directly ; but it does not need a great stretch of imagination
to see Sequanian caravans moving from the Jura to the Pyrenees
with their merchandise of wrought iron and sheaves of long-bladed
gaesa.

(I) The word * Sigynna ' is the name of a type of spear in
Cyprm. We are still very far from having shown the relevance of
the last statement of Herodotus about the Sigynnae ; that the
Cypriotes call their spears by this name. But here the evidence is
clearer. In the first place, there is no doubt about the Cypriote usage.
Aristotle \ in the fourth century, writes that the word triywov is * a
regular expression in Cyprus, but with us ' (that is, in the Aegean) * it
is a provincialism '. Here we have the word spreading from Cyprus
into the rest of Greece : and if the word, then presumably cilso the

^ Aristotle, PoeticSj 21 to <rty\/vav, Kvwplot^ fih icvpioi^, ^^pAV Sc y^^Tra.

272

THE SIGYNNAE OF HERODOTUS

thing : and we know enough about the repute of the Cypriote iron-
working in Hellenic times to accept this reasoning with some
confidence.

Later on, the word aiyvwa and its variants did actually become
fairly common ; but its later meaning diverged from the earlier ; and
a late but learned commentator notes that the divergence was two-
fold J In its ordinary acceptation, the term sigynna included any
spear which was * scraped to a fine point ', and this (though the word
used— ^oToj/ — properly applies to carpentry) exactly expresses the
pencil-like sharpness of the pilum and all similar derivatives of the
narrow-bladed type of spear/^ But 'in Herodotus'— and it is only in
the passage under discussion that Herodotus ever uses the word —
sigynmi signifies * the (well-known) th rowing-spear, whoUy of iron ' ;
almost exactly the phrase by which we have seen that Hesychius
interprets the word yaia-ov^ namely, as a * spit^like spear, wholly of
iron '. It is clear from this that there remained, at least in learned
circles, a definite tradition as to the meaning which the word had in
Cyprus in the fifth century, two centuries and a half before the
Roman adoption of the Spanish gaesum and more than two centuries
before the date of the Gaesate tunmUus,

Was there, however, any such weapon in use in Hellenic
Cyprus ? Here we are slightly better off than we were in the case
of Spain : for several examples are known of a type of spear-head
pecuhar to this island, dating from the fifth century and earlier, and
presenting resemblances to the Roman pilum so close that at the time
of their discovery they were regarded with some suspicion as the
result of some contamination of the find ; several of them are in the
C3T>rus Museum,^ they come from Tamassos, and were excavated by
Dr. Ohnefalsch-Richter in 1885. They are not complete, and the
longest fragment only measures 0-466 metre : their diameter is not
stated in the printed catalogue, but as nearly as I can remember, it
was between one and two centimetres. Another example is in th^
Fitzwilliam Museum at Cambridge: it forms part of a collection formed
by Sir Henry Bulwer, in Cyprus, about the time of the excavations at

* Schol, Plato, p* 384 <rty\;iT09 6* iari fwrroy 5o/iv, wop 'HpoSortii Sc to oAoertiSi/por
ojcovrtoi^.

* We have already seen, moreover, how Polybius illustrates the lighter
variety oipUum by comparing it with a '^ifivm^ {aiyivij) of moderate aize\

* Myres and Ohnefalsch-Ricbter, Cyprtis Museum Catalogue (Oxford, 1899),
No. a926 £r.

AN ETHNOLOGICAL PROBLEM

273

Tamassos. Some of the objects in that collection bear tomb-marks
showing that they are duplicates from Dr. Ohnefalsch-Richter's
Beries ; and though there is no documentary evidence on this point,
I have little doubt that this weapon is a dupUcate of G, 31. G, 3926 flfl

In the Cypnis Museum Catalogue^ Nos. 3926 flF, are described as
'long cylindrical spits', an unconscious tribute to Hesychiuss
phrase ifi^oXiou oXoaihyjpov ; and when working at the Catalogue
I had no better theory to advance. But the Cambridge example
suggests difficulties : it is better preserved, and shows traces of a
socket. Further, under date September 8, 1907, Dr, Ohnefalsch'
Kichter writes that several specimens were found, with well-preserved
sockets, in grave 12 at Tamassos in 1889, along with a fine iron
sword. The longest measured '735 m. Others ai-e in the BerUn
Museum.

The circumstances of discovery preclude the idea that these
are Roman pila which have gone astray. The Roman occupation of
Cyprus in 68 b. c. was effected quite peacefully. Moreover, a tomb
from Kition, now in the Grassi Museum at Leipzig, contains
examples of the same type in btmuse (length 82-90 m.) together
with a bronze spear-head of sub-Mycenean lanceolate type, and
sub-Mycenean pottery, probably of at least the ninth century.

We may, therefore, take it as certain that there existed in
Cyprus, in the ninth century and earlier, a type of iron spear almost
indistinguishable from the Roman pilmn, and known in the fifth
century by the name of the Sigynnae, But how did it reach Cyprus ?

Our information about the Early Iron Age of Cyprus is not so
copious as the extensive excavations would lead us to expect. Two
iron spear-heads, now in the Cyprus Museum, fi'oni Tamassos (3921)
and Amathus (3922), merely show translation of the normal late
Mycenean bronze type^ into the new metal. One-edged knives, also
descendants of a late Mycenean type, ai-e fairly common at
Katydata and other sites of this period. And one very fine iron
sword from Kurion has been pubUshed and discussed by Dr, Naue
and by Dn Arthur Evans, -

The type of sword to which this example belongs is unknown

' Well illustrated by C. M. C, No. aSOl, and Brit, Mm. CaL Bronses, No.
2770, both from Amathus,

* Naue, Die vmr6mmhen Schwcrter (Mtlnchen* 1903), p. 25, n., PL VI. 4. Evans.
'Mycenaean Cypms/ &c. , in Jounu Anthmp, Institute, xxx, p, 218, fig, 15, Compare
his roview of Dr, Naue 6 book in MaUj 1904, 24,

THE mOTWSAR OF HEBODOmS

taattninteiltotiie

and to t^^xA it wm the smord of

^UMTfnm. as HerodotoB

«t mU events^ b j that
isftfieHallrtattfegioii; SDrtheHaDBtBttswooi, wDr.NjneliMriKyini
in die diraet line of rmrcfnoioii to iins Itala-
^l^pell'. Ta Ifaid'l^ n'mtndiiMd iBto Cypnis—
wmA into Cjpn» alone (m Iv aa we know) in all tlie Leyant — as
the regular form of Early Iron Age ^wocd m^ tlie refiare , a ^
aoaBDeBtary on the dMtributi on of tfie aiiynna type of apeai;

Aa to the mode of ii< . — mniratio n befcipeen the Hallstatt
and Cjpnai two Tiena m^ posabfe. The Aatiibiition of the awotd*
type tmggfrti i it at first a^ht» diqiersal by mema of the ^driatir
a e a wa ya : and certain pcnnts in the aeries of fibnlae and other
bfime work, and also in the painted pottery c^ the early Iron A^
in the Lower Aegean and in C3rpnis, seem to support this view.
We cannot forget, either, that the Taphian traders of the Odyssey—
who put in at Ithaca with a cargo of iron, on their way to get
brr^ize from Temeae in Sooth Italy (or was it Tamassoe in Cypnia ?) —
mofgoA alao as fiu* aa Sidon, and eonaequently within sight of Cypni&>

But there is another possible connexion. We have seen
aheady how Siginni, with their trousers and ponies, were
recognizable by Strabo in the CancsBOB^ In an area which, as
IL Chantre has shown,^ shared, at least marginally, in the culture
of the Early Iron Age of Europe ; how a great iroa culture existed
m early Hellenic times among the Pontic Chalybes, and was
exporting its wares tm Ionia to the ports of the Syrian coast ; and
how the name of the Adriatic VEneti recurs in Homeric Paphla-
gonja, and seems to be presumed there stiU, in the phrase employed

* Homer, Odfueff L 181 (Tenme), xt. 427 (Sidon). The eYidence of Stnbo,
Ml^ is iteoiiglj in &voor of the Italko Teme

AN ETHNOLOGICAL PROBLEM

275

by Herodotus. All this tallies wholly with the description of
Apollonius Rliodius, who places liis Daniibiaii Sigynnae in Moesia
alongside * Scythians mixed with Thracians *, at the junction, that is,
between the stretmi of immigrants south-eastwards from Central
Europe into the Aegean/ and the south-westward stream of nomads
out of the grassland of South Kussia.

Even so, we are not very close to Cyprus ; and in the present
state of our knowledge of the early histoiy and culture of Cihcia
and South Cappadociaj we have to fall back on indirect evidence
once more. On the one hand there is the clear fact that the collapse
of the civihzation of the Aegean Bronze Age left Cypi-ua isolated
within a Levant wliicli was rapidly being Semitized by nascent
Phoenicia ; and tliat, isolated thus, Cyprus began to draw largely
from a culture which was not Semitic, and was appaiently Anatolian ; ^
only ceasing to draw from it when the renascence of the Aegean
permitted the anti-Semite factions in Cyprus to rejoin hands with
their Ionic kinsmen* The piratical attack on Sai'gon s exposed sea^
flank in Cihcia in 718 b,c. very nearly marks an overlap of the two
phases ; for Sargon clearly regarded this as a sea-raid, a concerted
movement by allies of his foes in Cappadocia,

Under these eii*cumstances there is no difficulty in supposing
that Cyprus acquired important elements in its Iron Age culture from
an AnatoHan source. Whether among these elements we are to
include any part of its iron metallurgy or its types of weapons must
still be uncertain, as long as what I may call, for short, the ' Taphian
route ' remains a possible alternative ; and on this point excavation
alone can decide. Wliat is chiefly wanted is (1) accurate evidence
about even one of the early sites in Cilicia ; (2) surface exploration
in Pontus, to test on the spot the Greek beUefs about the metallurgy
of the Chalybes ; and (3) thorough excavation of the Late Minoan
settlement at Tarentum, which, originating as it does in a period
wliich must be closely contemporary with the Mycenean coloniza-
tion of Cyprus, lasts on like the Mycenean settlements in Cyprus
in a state of suspense ; falls into a decadence so nearly identical
as to raise the suspicion that some sort of direct communication

* This stream also is traceable backwards, as we have seen, at all events as far
as the Homeric Age*

'-' This is not the place to maintain this thesis in detail ; the more obvious
materials are discussed in my paper, * On the Early Pot- Fabrics of Asia Minor/ in
Jaum. AtUhrop, Tmtituk, xxxiii, S67 ff ; esp. p. 393.

276

THE SIGYNNAE OF HERODOTUS

was maintained throughout, and finally revives, like Cypiiis, into
a new local Hellenism, and becomes a great outpost of Greek trade
and industry.

I conclude by summing up, briefly and dogmatically, the results
to which the evidence here collected seems to point. The Sigynnae
represent a people widely spread in the Danubian basin in the fifth
century b.c, from opposite Thrace to the head of the Adriatic, They
were trading across the Rhone valley westwards at that time ; and
Sigimii are found, later, to have estabhshed themselves in the
Caucasus among peoples of kindred culture. They are very likely
identical with the Sequani of the first century ac, and their culture
is that of the trouser-wearing, pony-driving region in which Herodotus
places them. His scepticism as to their Median ancestry is justified ;
if there was any connexion between them and the trouser-wearing
conquerors of Media, it was probably the other way round The clue
supphed by Herodotus as to their westward trade, together with the
form of their name and its derivatives, connects them witli the iron*
working culture of Hallstatt and later of La Tene, Tliis culture,
intrusive both into Spain and into Italy, gave rise inter alia to
a highly specialized series of narrow-bladed, and eventually wingless,
throwing-spears, which culminate in the gaesum and the pihim. Then,
the other clue supplied by Herodotus, as to the name of a spear-type
in Cyprus, enables us to use these West-Mediterranean materials to
identify extant examples of this^i^i/tma-spear ; to attribute these early
Cypriote pila to a northern origin, remote but perhaps direct ; and
so to get some light on the nature of the process by which iron-
working was introduced into Asia Minor and the Levant, and in
particular on the circmnstances under which the Caucasian Sigimii
may be supposed to have effected their migration.

I trust that this attempt to comment at some length on a typical

* Herodotean digression ', inadequate though the materials are — and
though the treatment therefore must necessarily be — has at least been
sufficient to show that modern archaeology, and still more modern
ethnology, has much to contribute to the interpretation of an author
who, according to the light of his age^ set himself as his life-work to

* save from obhvion the deeds of men whether Greek or barbarian *,
and, more than this, to discover * from what cause * arose the eternal
struggle between East and West in the Mediterranean.

A MUSEUM OF ANTHROPOLOGY

By C. H. READ, F.S.A.

Keeper op the DEFAEiidENT of Ethnography in the British

Museum

In a volume of this kind it is hardly necessary to enlarge upon
the value and importance of a museum of anthropology. I have
on several occasions, at the British Association and at the Anthro-
pological Institute, insisted that the creation of such a museum is
a duty that England owes to herself, that it would render definite
service to the State, and that it could be formed at less cost by
England than by any other country. While England is lethai'gic,
(Jermany has created an ethnographical collection in Berlin ten
times the size of tliat in the British Museum, and equal in im-
portance to all the similar collections in England put together.
How posterity will regard such supineness one can hardly foresee,
but there will surely be some regret at the loss of opportunity to
secure a tangible record of the conditions of the many primitive
races that still exist, though they are inevitably changing.

For practical reasons, as well as for scientific completeness, the
collections of physical anthropology should fomi part of the same
institution as those relating to etlmography. They are often apt to
overlap in their practical aspect, while the study of the divisions of
the human race must necessarily include both. To a certain extent,
moreover, such an arrangement would be an economical one as re-
gards the staff. By far the greater part of the exhibition space would
of com"se be given up to the ethnographical section. Collections of
this kind need far more space for their proper display than is devoted
to them in any museum in Europe. Perhaps the bast equipped in
this respect is the American Museum in New York, where certain
sections are shown in an ideal way. Adequate exhibition is a sure
step towards popularity, and no public museum can hope to exist
without the appreciation of a large pubHc apart from specialists.
Such a policy not only attracts donations and support in other

278

A MUSEUM OF ANTHROPOLOGY

directions, but it has an effect of even more lasting value, that
of creating new schools of students, new lines of study, and of
constantly imparting a new impulse to the whole machine. Tlias
a great area of exhibition space must of necessity be devoted to
ethnography. The differences* in the physical character of the
human skeleton are in the main more difficult for the lay mind to
follow, and for this reason there is less need for devoting much
space to their display to tlie general public, who would naturally
take a greater interest in the published results. There are, in fact,
two distinct reasons for exliibiting objects in a museum. Tlie first,
and more serious one, is exhibition for the purposes of instruction,
the second is for the edification and tlie entertainment of the un-
learned majority, and, for the reasons just stated, both must be kept
well in mind if the museum is to meet with public support. Thus
a very large part of any scientific museum must be given up to the
display of collections which, if the serious work of the institution
were alone considered, might in the main be aiTanged in drawers or
cupboaids with a great saving of space.

Tlie foregoing paragraphs deal with the conditions of a museum
where the material with which it deals has already been gathered ;
but it may be well to add a few words pointing out the principal
ways in which such a collection may even now be formed, and the
help that may be got towards bringing it into existence.

Unfortunately, in a sense, it may be assumed that no Grovem-
ment of a great empire hke our own will of its own motion take
steps in any such direction. It is only in smaller countries that time
can bo found for a Government to foster science or learning for its
own sake, as a factor in the future well-being of the people. Ex-
ceptions are occasionally found where the ruler or a member of his
family may take a far-sighted \'iew of the ultimate public good, and
by the exercise of a vigorous poi-sonality, produce the same results,
with the aid of the public purse, as if the Government itself had taken
action. Such a condition is, however, so rare that it may practically
be left out of account. It does, however, exist, and where it exists,
produces no small effect.

In England, almost all great enterprises, from the possession
of the Indian Empire downwards, have had their origin in tlie
adventurous spirit of private individuals. Their small beginnings
have at length attained to such greatness as to force tlie attention
and even sympathy of the Government, and in tliis way the

A MUSEUM OF ANTHROPOLOGY

279

empire has been built. Such being the case in the past^ it is wise
not to look to the central government for more than a * grant in aid ',
and a few benevolent platitudes^ until such time as the scheme has
so ripened as to command attention. There is, however, one way in
which the Government machine may be made to contribute towards
the success of an anthropological museum. The Foreign Office, the
Colonial Office, and the Intelligence branch of the War Office and
the Admiralty, have in their employ, or under their control, a vast
army of trained intelligent officers, who, if they will, can very
materially help in such a work. The one condition necessary, how-
ever, is that the work in this direction that such officers perform
should be carried on with the approval and sjonpathy of their
superiors at home, and that the work, adequately performed, should
be regarded as creditable by the central office so long as it be done
without interfering with their primary duties.

Assuming, then, that the Colonial Office, for instance, authorizes
the officers of the Anthropological Museum to invite the co-operation
of Colonial Governments and their subordinate officers in the forma-
tion of such collection in London, it would be found that of such
officers about one in ten would be eager to employ his intelligence
in leisure time in such work. If, therefore, one-tenth of the
forces abroad from the above-mentioned departments could be in-
duced to promise their help, what might not be done in the way of
collecting in all branches of anthropology ? One incidental result
would ensue. We should then have in London a teaching agent
of inestimable value, viz, a place where Enghsh people at home could
form an idea of what the British Empire means. The Indian and
Colonial Exhibition was the only thing that at all answered the
question, but how many ordinary English people remember it now,
after a lapse of twenty years, except in the mistiest way ? I am not
forgetful of the existence of the Imperial Institute. It is now doing
good work, and is becoming a bond between the Colonies and
England : but a series of lamentable mischances in its early life has
irrevocably shorn it of much of its glory and comprehensive chamcter.

So much, however, for government aid. There are fortunately
other helpers. The great trading companies, whose enterprises take
them among uncivilized races, are usually very ready to lend their aid
in scientific matters, while many of the individuals governing their
affairs are not infrequently men of great breadth of view and occasion-
ally ready to help in forwarding the cause out of theii* own pockets.

280

A MUSEUM OF ANTHROPOLOGY

Where one adds to these the reUgious and scientific missions, all of
them again supported by private fiinds, it would seem that there
need be no fear as to the success of such an obviously necessary
institution as an anthropological museum*

The collections having been got together, the next question is
the best system of arrangement, and what kind of a building is the
best for the purpose. As to the first, it is generally agreed that for
a general ethnographical collection the geographical system is the
most practical and the most advantageous ; the most practical
because both the student and the public know more readily where
to look for a given object, and the most advantageous because the
geographical plan, apart fi*om its more apparent benefits, begets a
kind of patriotism or local pride in the workers in particular areas.
A man, for instance, who has spent the greater part of his life in
Borneo, will generally find a special satisfaction in seeing the collec*
tions he has made ananged as a unit in a museum. Wliereas, if
they were dispersed under some plan of showing the development of
special crafts, his satisfaction would take a modified form, Anotlier
objection to the evolutionary anangement is that the student camiot
gain an idea of the state of culture of any particular people.

In every such museum, however, the ideal system would be to
have practically a duplicate collection in many branches, in order to
nourish subsidiary series demonstrating the evolution or distribution
of particular types. In very many cases, however, such demonstra-
tions can be as well or even better furnished by dramngs or
photographs, especially where the originals are at hand for reference
and more effective demonstration.

I speak here only of the ethnographical specimens, for the ex»
hibited portion of the physical anthropology would be comparatively
simple to deal with and relatively small.

There are doubtless many types of building that would serve the
purpose of an anthropological museum, but I think the most practical
ground-plan is that of two oblongs, forming galleries, one within
another. The outer one and part of the iimer would contain the
ethnography, the remaining portion of the inner one the physical
section. The two would be connected at inter\^als by rooms to form
workrooms for the staff, or small galleries containing the subsidiary
series spoken of above. The parallel galleries have, moreover, a defi-
nite and practical advantage, inasmuch as they allow of the closing
to the pubhc of any section of any gallery without stopping the cir-

A MUSEUM OP ANTHEOPOLOGY

281

culation. At the beginning of the closed part of the outer gallery,
for instance, the public would be diverted into the inner one, and
emerge again into the outer gallery at the end of the closed portion.
Experience has shown that such a plan has great practical value in
dealiiig with the re-arrangements which are constantly necessary.

i R A R Y

Caller y

O A L L E n Y

ENTRANCE

The place here advocated is, of course, capable of infinite modi-
fication while retaining the general principle, which is the only
essential part. In a hall or vestibule might well lie placed an index
series to give the visitor a general idea of the meaning of the
collections. Further, in an annex of some kind should be placed
&, library, mthout wliich no museum can pretend to be complete.

This brings me to another part of the ideal museum scheme ;
the office tor the collection of anthropological information.

It would seem an obvious matter that in England, and under
the control and superintendence of the Government, there should
exist a central office, where all information relating to native laws
and customs should be gathered together, and so arranged as to be
useful to the officers proceeding to take over the government of a

282

A MUSEUM OF ANTHROPOLOGY

particular district. No adventurous trader would think of putting
responsibility among uncivilized peoples upon a young man fresh
from school or college without giving him a chance of learning the
essential conditions of his district. He would have to learn the
language to some extent, to master the system of trade, and to
familiaiize himself with such of the customs and religious beliefs as
relate to daily life. Such a preparation in commerce is considered
an essential preliminary, whereas in dealing with the government
of primitive peoples any such training is held to be almost entirely
neghgible. It must be confessed that owing to some sympathetic
qualities which seem innate in the Anglo-Saxon blood our hetero-
geneous empire has grown up with singularly few serious blunders,
in spite of the haphazard empirical methods that have ruled it. But
the relations of empires and peoples are inevitably undergoing a
change from the old rule of thumb system that has hitherto suflBcedj
and more precise, accurate, and strictly scientific methods will be the
rule in the near future. To attain to these, wide and detailed
knowledge of the domestic conditions, of the religions or superstitions,
of the daily habits, and even of the prejudices, of every native com-
munity has to be obtained and used as the foundation for the new
departure* Many a native revolt, involving the loss of much money
and many valuable lives^ could have been avoided, if only the white
man had been able to enter into the ideas of his brown or black
neighbour.

For these reasons, therefore, quite apart from the gain to science,
some such bureau of anthropological inquiry should exist in England.
Some years ago I formulated such a scheme before the British
Association, with the suggestion that it should be under the auspices
and superintendence of the Colonial Office. Eventually, however,
it was decided that it should be estabUshed in my department at the
British Museum. There I have done the best in my power, but I
realized from the first that it could not attain to anytliing like per-
fection without a much larger expenditure than could be allocated to
a small department. By the circulatioii of copies of the * Notes and
Queries * issued by the British Association, and also of special
* questionnaires ' prepared for special districts, a good deal of in-
formation has been obtained, and no small amount of incidental
benefit has accrued to the museum. One unexpected result also has
come about. The questions issued have been used as the fouudsr
tion of published works by officials of various protectorates — a result

A MUSEUM OF ANTHEOPOLOGY

283

by no means to be deplored, though it does not serve the immediate
purpose of the bureau.

It would no doubt be an ideal system to deal with the informa-
tion gained by pubHshing it in the form of monographs, as is done by
the Smithsonian Institution at Wasliington. But in America the
field, though large, is at any rate Umited^ and even then, is no doubt
a costly affair. With us the amount of material would be ten times
as greatj and adequate publication would mean a great outlay. The
Soci6te beige de Sociologie also has formulated a scheme which has
been set out in a pamplilet entitled *Enqu^te ethnographique et socio-
logique Bur les Peuples de civOisation inferieure' (Brussels, 1906),
and M, Joseph Halkin presented a report to the Congres Interna-
tional d'expansion economique mondiale de MonSj under the title
* Une Enquete ethnographique mondiale ' (Liege, 1905). Hie former
has a purely scientific pui-pose, while the latter aims more at ad-
vantages to trade. Both, however, contain a good deal of common
sense and useful suggestionSj which it would be well to bear in
mind in starting any sui'vey for scientific purposes. The question-
naire in the first-named of the two pampUets is, like the report,
from the hand of M, HaUdn, and is divided in a way easy to com-
prehend and follow. The divisions are as follows : 1, Vic niaterielle ;
2. Vie familiale ; 3. Vie religieuse ; 4. Vie intellectueUe ; 5. Vie
sociale ; and 6, Caracteres anthi'opologiques, divided into *somatiques'
and ' physiologiques \ Under each of these heads are given sugges-
tions for detailed questions which are, of course, capable of infinite
expansion and variety according to the needs of special districts.
Tlie physical section especially is too hmited in its range in this
Belgian scheme.

If a country like Belgium finds that inquiries of the kind are
justified with such a limited area as it possesses for their application,
how much the more are they necessary in England But to be of
real use the bureau must be a definite institution, having intimate
and sjnaipathetic relations with the various branches of the Govern-
ment, it must have a director who should be a trained scientific man,
and an adequate staff to assist him. Under these conditions, a
bureau of anthropology for the British Empire would be an eminently
useful branch of the public service.

THE NINE WITCHES OF GLOUCESTER

By sir JOHN RHYS,

Pbofessoe of Celtic,

Principal of Jesus College, Oxtokd

The passages bearing on the Witches occur in the story of
Peredur. That knight in his wanderings came once on a time to
a castle on a mountain : then follows an episode which will be found
in the Oxford volume containing ' Tlie Mabinogion and Other Welsh
Tales', pp. 210, 211, to the following effect — *And when Peredur
canio into the hall there was a fine tall woman sitting in a chair,
with many handmaids around her ; and the lady welcomed him.
When they found it time they went to eat : when they had done
eating, she said, ** It were well for thee, prince, to go to sleep
elsewhere/' ** Why should I not be allowed to sleep here ? " said
Peredun "There are nine witches here, my souV said she, *' of the
Witches of Gloucester, together with their father and their mother ;
and by daybreak it is not more Ukely that we escape than that we
are killed by them/ They have conquered the dominion and laid it
waste except this one house." ** Well,'* said Peredur, *' it is here we
shall be to-night, and if trouble overtakes you I shall, if I can, be of
service to you : at any rate it will not be the contrary of service on
my part/' They went to sleep, and at the dawn of day Peredur
heard a terrible cry. He rose quickly in his shirt and trousers,
with his sword hanging round liis neck, and he wont out What he
beheld was a witch overtaking a watclmian, who was crying out.
Peredur made for the witch and dealt her such a sword-stroke on
her head that it flattened out her helmet and headgear like a dish
on her head. '* Thy protection, fair Peredur son of Efrawg, and God s
protection ! " ^* Wliy, hag, knowest thou that I am Peredur? " '* It is
destiny and vision that we suffer at tliy hands, and that thou on thy
part obtain a horse and arms fi*om me. Under my charge shalt
thou learn to be a knight and to handle thy weapons/' ^* Thus," said

* Lady Charlotte Gnest {MaL^ L 323) has mistranslated this sentence — 'and
unless we can make our escape before daybreak, we shall be slain.' M, Loth's
translation is also wrong : see Les Mabitio^ion, ii* 09.

286

THE NINE WITCHES OF GLOUCESTER

Peredur, '* shalt thou have protection, namely, if thou give thy troth
never to do any wrong in the dominion of the Comitess." Peredur took
security on that score, and, with the leave of the Countess, he set off
with the witch to the Court of the Witches. There he abode three
weeks on end : then he chose Inm a horse and arms, and thereupon
he set oflf.'

The sequel at the end of the Peredur^ pp. 242, 248, connects
the Witches with the Grail Stories : see Nutt's Studies on the Legend
of the Holy Grail^ pp. 101, 139, 143, 156. Tlie Peredur passage runs
to the following effect : — ^ *' Lord/ said the young maUy " it was I who
came in the shape of the black maiden to Arthur s Court ; and when
thou didst throw the chessboard, and when thou didst kill the black
man of Ysbidinongyl, and when thou slowest the stag, and wert fight-
ing with the man from the Llech. It was I also that brought the head
bleeding on the dish, and the spear wliich had a stream of blood
trickling along the shaft from the point to the heel. The head was
thy cousin's, and him the Witches of Gloucester had killed ; it was
they also that lamed thy uncle. I am thy cousin, and it has been
predicted that thou wilt take vengeance on them;" Peredur and
Gwalchmai took counsel together and sent to Arthur and his house-
hold to ask him to come and attack the Witches. They began
to fight with the Witches, and a man of Arthur a was slain in the
presence of Peredur by one of the Witches, though Peredur forbade
her. Tlie second time the Witch slew a man in Peredur s presence,
though he forbad her again. And the tluid time the Witch slew a
man in Peredur's presence, whereupon Peredur drew his sword and
struck the Witch on the top of her helmet so that he split the helmet
and all the armour together with the head into two halves. She
uttered a cry and bade the other Witches flee, saying that it was
Peredur who had leaint knighthood with them, the man who was
destined to slay them. Then Arthur and his household struck in
among the Witches, and the Witches of Gloucester were aU slain.'

Here in both passages the word for witch or sorceress is gimdon^
plural gwklonotj which would be in modem spelling gwiddon^ gwidd-
onod. Tlie singular occum also in two passages in the Story of
KuUiwck mid Olwen, pp. 123, and 141, 142. Both relate to the
necessity for Arthur of procuring the blood of y Widon Ordu^ ' the very
black witch/ daughter of y Widon Orwen^ ' the very white witch,* from
Pennant Govud on the confines of Hell ; and the second of them
runs thus : — Arthur set out towards the North, and came to the place

THE NINE WITCHES OF GLOUCESTER

287

where the Ixag had her cave. Gwyn, son of Nudd, and Gw3rthyr,
son of Greidawl, counselled that Kacmwri and Hygwydd, his brother,
should be let go to fight with the hag. As they got inside the cave
the hag was beforehand with them : she seized Hygwydd by the
hair of his head, and struck him down under hen And Kacmwri
seized hold of her, and pulle<:l her down from off Hygwydd. She
turned round upon Kacmwri : she belaboured them both, she dis-
armed them and drove them out uttering cries of dire distress.'
Arthur grew wroth at the sight of his two servants all but killed,
and he tried to make a dash into the cave, whereupon Gwyn and
Gwythyr said to him : *' We think it neither nice nor amusing to see
thee engaged in a scratclung match with a hag : let Hiramreu and
Hireiddil go into the cave." They went, but though the trouble of
the previous two was great, greater was the trouble of the latter
two. Heaven knows that not one of the four could move from the
place, except for their being placed on the back of Llamrai, Arthur's
mare. Then Arthur made a dash into the opening of the cave,
and cast Carnwennan, his knife, at the hag, and hit her in the middle
so that she fell asunder in two hulks. And Caw of Prydain took
the witch's blood and kept it in his charge/

Both in the Kidhtvch and the Peredur the witch is addressed as
gtcntchj * hag ' ; but in the former the hag does not appear to have
any armour : she uses her hands, and Gvrjrn and Gw3rthyr evidently
feared to see her using her nails in a conflict mth Arthur. It
was otherwise with the Witches of Gloucester : they worc helmets,
and apparently other armour. The Peredur story associates prophecy
and prediction with them, though not in such terms as to prove
beyond doubt that they themselves were the authors of any such
prediction any more than Peredur's cousin, who also knew of the
prediction that the former was to destroy the Witches.

The word gmddon is used by our fouiteenth-century troubadour
D. ab Gwilym in his cyivijdd^ clix, line 62 —

Cryglais gividdon utemn croglatK
* The hoarse voice of a witch in a noose/

Dr. Davies in his Dictionary has gwiddon explained as ' malefica,
saga, gigas faem '. The modern tendency in books is to make the word
into gtviddanj which Pughe in his Dictionaiy explains as ' A hag ;

* Lady Ch, Guest renders it *with kicks and with cuffs', and M. Loth
translates to the same effect {Les MaL, 1. 282) ^ but 1 prefer the above.

288

THE NINE WITCHES OF GLOUCESTER

a witch ; a sorceress ; a giantess,' together with a derivative gwidd-
aneSj *a witch/ This last is used by the Bardd Cwsg with y for i, and
he has also the simpler gwijddan : see Prof. J. Moiris Jones's edition,
pp. 61, 55, 189. On the other hand I am familiar in North
Cardiganshire with a stream called Nant widdon, in Djffryn Castell ;
it falls into the Castell river on the left-hand side as one proceeds
from the Inn to Eisteddfa Gurig. The name is sometimes pro-
nounced Nant iciddol; but I presume Nant widdon to be more
correctj and to stand for Nant y Widdon, * the Brook of the Witch.'
The outcome of tliis is that the oldest and best attested spelling is
gtviddon and not gwlddan or givgddau : so the origin and history of
the word are left all the harder to explain,

On reverting to the passages in the Kulhwch and the Peredur
one may take it that the gtviddmi in the former was an ordinary
witch or ogi'ess Uving alone in a cave, wliile Chmddonod Caerhgw in
the Peredur were warrior women or amarons, who were regularly
armed for both oflfensive and defensive action ; and they had a
settled home called Llgs y Gwiddonod^ * the Witches' Court,* where
they had with them their father and their mother. There also they
taught the young men who came to them feats of arms and all that
was understood by chivalry in the society to which they belonged.
That society was evidently the same to which belonged Scathach
figuring in the Cuchulainn stories, which I have summarized in my
Celtk Heathendom^ pp. 460-5, 480, 481.^ Scathach is also repre-
sented as having her court somewhere in Britain, though that state-
ment is somewhat obscured by the account of the access to her
court being taken from the stock description of the Bridge of the
Dead. For it is evident that these stories are composed of two
elements^ the mjrtliical and the historical. The female warrior is
not mjrtliical, but belongs to early Irish history : the employment of
women as warriors was only discontinued in the last years of the
seventh century. Tliis advance in humanity is usually associated
with the name of Adamnan as the Law of Adamnan, otherwise

* The whole requires revision, hut I will only mention here the place called
in Irish, p. 451, Olefin n'GdibtJiechj * the Sorrowful Glenn,' where gdibtfie^h is the
adjective from gdbudj which is the same word as Welsh gouul^ now pronounced
gofid, though the spellings gofyd and gofuil also occur : see Br. Da vies, s.v. At all
events the Welsh * Pennant Go\^d * on the confines of Hell comes sufficiently near
Glenn n'GdiUhck to testify to the common origin of the portion in question of tlie
Welsh and Irish stories*

THE NINE WITCHES OF GLOUCESTER

289

called Cain or Mecht Adamndin^ and Lex InnocentiumJ Accordingly
there need be nothing essentially non-historical in the whole incident
where the Witches of Gloucester figure except the predictions. It
all points to a Goidelic head quarters somewhere in the neighbour-
hood of Gloucester. Various other facts seem to point in the same
direction, that is to say, to the land of the ancient Silures ; among
them may be mentioned the comparatively large number of Goidelic
names of places and persons within the ancient boundaries of the
diocese of Llandaff as represented in the Liber Landavensis. Attention
has been called to the most remarkable of them in a paper of mine
entitled *The Goidels in Wales* : see the Journal of the Archaeologia
Cambremis^ 1895, pp. 18 et seq.

The names imply a considerable invasion of the country west of
the Severn estuary by Goidels sailing up that water ; and the date
may be guessed to have been from the second century to the sixth.
Possibly it was a continuation of the movements of the D6ssi, who
occupied Demetia in the latter part of the third century : see the
Ar, Cam, Journal^ 1892, pp. 6&, 66. Here may be noticed Geoffrey
of Monmouth's mention, in his Eist Begum Britanniae^ iii. 10, of
Kaerosc, otherwise Caerleon or Urbs Legionis as * metropolis Demetiae*
before the advent of the Romans. We need not imitate liim in
putting Caerleon back to such an early date, but a point to be noticed
is his having got somewhere the idea of extending Demetia so as
to include the land of the ancient Silures (Geoflfrey, iv, 15), I gather
that the authority which he followed made the Goidels of Demetia
and those of Siluria the same people, with their centre of gravity
in the latter rather than in the former. Further, the mabinogi of
PwyU, Prince of Dyfed or Demetia, gives that prince a liegeman io
Teimyon Twryf Viiant, prince of Gwent Iscoed or Nether Went,
in what is now Monmouthshire. This occurs towards the end of
the mabinogi (Mak, p. 22), which closes with a few words about
PwyU's son and successor, Pryderi, to the effect that the latter, when it
occurred to him to marry, chose as his wife Kicva, daughter of Gwyn

* The following are the principal references in point — I owe most of them
to a friend's kindness : Reeves, Adamnan's Vita Columbae, pp* I, Hit, 179 ; Felirc
Oi«^5^^ 3ept 22, and p. 211 of Stokes's ed. 2; Cdin Adamndin (ed. K.Meyer),
Oxford, 1905, pp. 6 et aeq, ; Windisch's Irischt Texte, p, 19S, 11. 6-8 \ * Annals of
Ulster * {vol. i, ed. Henneaay), a. d. 696 (pp* 144-7) ; O'Donovan, * Annala of the
Four Masters/ a,d* 703, note u with references to the Speckled Book, fo. 38^, and
the Book of Lecan, fo. 166, p. a. coL 4 ; and D'Arbois de Jubainville'a FamiUe
CeUiquef pp, 81-3.

290

THE NINE WITCHES OF GLOUCESTER

Gohoyw, son of Glojw Gwallt-Iydan, son of Casnar Wkdic (*C. dtix ')
of the rulers of this island. Now Kicva is a decidedly Irish name,
which figures in the early and legendary portion of Irish history :
see the *Four Masters' Annals of the Kingdom of Ireland ', a.il 2520,
Kicva's father is called in the mabinogi by a name which may be the
Welsh rendering of some such Irish designation as Finn Fos^im^
but I am more inclined to regard Gohoyw as a coiTupt version of
Gloyu\ As soon as the two stood there side by side the latter was
provided with the epithet GivaUi-Iydan^ 'Wide-haired/ elsewhere
(jhvaUtit\ * Long-haired/ and Crwhd4ydan^ perhaps * Widely-ruling \
This conjecture is borne out by the beginning of the third mabinogi,
that of Manawyddan^ son of Llyr, where Pryderi is made to describe
his queen simply as Kicva, daughter of Gwyn Gloyw* We gather
that the scribe felt readier to add at the end of the one mabinogi
than in the body of the text of the other : one is almost certain to
be right in giving the preference to the shorter text.

The question now is how Gwynn Gloyw is to be interpreted : at
first sight one might render it * White-shining' or the * White and
bright'; but more likely one should treat Chvynn, 'white/ as the whole
name, and regard it as equivalent to the Irish Finn, Then Gloyiv has
to be treated as the epithet, but that epithet admits of two explana-
tions : treated as an adjective we should have *G^vyn the Bright or
Shining', but otherwise and better it would be *Gwyn of Gloyw', that
is to say of Glevum or Gloucester, for Glevum must be ghyw in Welsh ;
or rather the Latin name was the neuter of the Brythonic adjective
glevos^ glevd^ glevo-n^ now gloyw or gloeu\ 'bright, shining/ Possibly
Glevum has superseded some such a longer form as Glevo-castra^ or
Glevo-dtifw-n^ * Shining fortress/ The way in which such names
have been treated is very well illustrated by the case of Caerfyrddin,
* Carmarthen/ which means the caer of Maridunon, that is the
fortress of the Sea-town ; but this was forgotten, and Maridunon
which became Myrddin was treated as a man's name, so that
the whole came to be interpreted as Myrddin's Caer. Here Caer
Loyw, that is the Caer of Glevtim, came to be Gloyw's Caer, with
Gloyiv made into a personal name : ' compare Geoflfrey, iv. 15, making
Gloyw either into Claudius or a son of his, 'Gloio duce/ According
to the text of the Pwyll with its Gloyw WaUt-hjdan^ Gloyw was

' Mr, Nicholson suggests to me that Arianrhod was likewise a place-name, an
Argentoratum in Britain. If bo^ perhaps^ the lady's original name underlies the
modern Elafif which recalls Eihknn, genitive of the name of the Irish Lug's mother.

THE NINE WITCHES OP GLOUCESTER

291

made into a man, who came to be treated as the ancestor of the
GoideHc family of invaclers to whom Kicva belonged ; and lest that
should not be good enough Gloyw is made to be of the line of duke
Casnar of the rulers of this island. His line ^ is mentioned by the
twelfth-century bard Cynddelw: see the Mytyrian Arch., i, 239, where
we read Ef oed lary o Un Gastiar^ * A bounteous one was he, of the
Une of Casnar/ In any case the lady with the Irish name Kicva is
associated with Glevum or Gloucester, and her home was possibly no
other than that which the Peredur calls Llys y Gmiddonodj ' the Court
of the Gwiddonod/

There is more to be said of the shadowy personage called Gloyw,
for the pedigree of Fernmail king of Buallt about the end of the
eighth century (Zimmer s Nennim Vindicatus, pp. 70-3) is made to
end with him in the Mistorki BriUonum with which the name of
Nennius is usually associated. See Mommsen's * Historia Brittonum
cum Additamentis Nennii ' in his Chronica Minora, iii, § 49 (pp, 192,
193), where the names in point run thus: — *filii Guorthigirn Guoi>
theneu, filii Guitaul, filii Guitolin, filii Glovi • . , qui aedificavit
urbem magnam super ripam fluniinis Sabrinae, quae vocatur Brit-
tannico sermone Cair Glovi, Saxonice autem Gloecester/ GuortMgim
is the regular form in Old Welsh of Vortigem's name, Vortigemios,
and in later Welsh it has successively become Gwriheym^ GwrtMyrtu
In the next place some of the MSS, have forms wliich point to
Guitolion as the preferable reading here, and I take Guitolin and
Guitolion to be distinct names derived from the Latin Vitalinus and
Vitaliafms respectively : see the Cymmrodor, 1906, pp. 72-4, Lastly
for Giovi one should restore Gloiu or Gloiv, and then one may turn

* Is it possible here that Ckisnar was originally an error for th© Latin
Cae^aTy meaning Claudius Caesar ? The numerous coins of that emperor found at
Gloucester seem to connect him with that city, which Geoffrey calls Kaer^lou
(Saii-Marte*s text, iv. 15) and interprets as named so perhaps after the Emperor, as if
G?ow or Gloyw could be an equivalent for the name Claudius, So the Red Book
translation of Geoffrey knows Claudius only as Gloyw (Oxford Bnits, pp. 94-7)^
while the versions published in the ^Myvyrian Arch, of Wales' (iL 187-94) call him
Gloetv Kemr and even Gloffw Casar, Cynddelw's metre does not require Cctsnar
or Cesar, nor does vnsnar, * indignation, wrath, ire * (Silvan Evans's Gciriudur, s, v.)
suit the sense. It is interesting to notice that *laiy o lin Gasnar* seems to
have suggested to the Kulhwch story-teller his *Llary m, Kasnar wledic' [Mah.,
p. 107), which appears in 'Khonabwy's Dream ' (If afe.^ p. IdO) with Kasnat for Kasriur,
It helps one to understand how the lists of characters in these tales were made up ;
and it is needless to say that Cynddelw 's llurif is not a proper name, but simply the
Latin adjective lunjus boiTowed and reduced to a monosyllable llari,

U 2

292

THE NINE WITCHES OF GLOUCESTER

the names back into Latin thus : ' filii Vortigernii, filii Vitalia, filii
Vitaliani, filii Glevi/ Vitaliani is the genitive on a bilingual stone
at Nevem, in Pembrokeshire, and it is not improbable that it
marked the tombstone of Vortigern's grandfather. Recently
another bilingual stone has been found there : see the Archaeohgia
Camhrenm Journal, 1907, pp. 81-92, 310.

It has already been suggested that the Irish invaders were the
D6ssi and tribes allied with them, and I have elsewhere expressed
my view that they at one time haunted the coasts from Kerry to
Kent. On tliis side of the Irish Sea they had perhaps their largest
settlement in Dyfed or Demetia, within which are the village and
landing-place of Nevern ; and in this connexion it is to be borne
in mind that Vortigern's name occurs more than once in the
Irish territory of the D68si, in the County of Waterford, whence
they crossed to Dyfed. Far east from Dyfed we come to the lower
valley of the Wye, where the Dessi were probably the founders of
the kingdom of Buallt over which the Fernmail already mentioned
reigned in later times. Moving further eastwards we seem to detect
the D6ssi by means of the Irish names in the lAher Landavemk ;
and on the Severn somewhere about Gloucester we find the home
of the Irish Princess Kicva ; not to mention that Fernmairs pedigree
associates his ancestry with Glevum itself The Severn Sea would
seem to have been one of the highways of Goidelic invaders, and one of
the Glamorgan places called Bamven was the Bannaventa from which
Professor Bury {Life of St Patrick^ pp. x, 17, 321) supposes Irish
rovers to have carried Patrick away to Ireland. Add to this, that on
the Somerset side an old inscription occurs marking the grave of a
Nepus Carataci^ a decidedly Goidelic description of the deceased, pmc-
tically meaning Carthach's uterine Sister's Son, according to a common
Irish formula. The coasts of Cornwall and Devon have their Ogam
inscriptions and bilinguals to show, and the series ends at present
with the Ogam stone found some years ago in the excavations on
the site of the Roman city of Calleva, now known as Silchester, in
Hampshire. To complete the series, and practically to establish the
view here suggested, one only requires an Ogam inscription or two
to be discovered in Sussex or Kent* The hypothesis fits also in other
respects : among other things it helps to understand Vortigern s
movements,^ and even to account to some extent for the incest in

* Why does Geoffrey call Vortigern consul Gewisseorum? Can the original have
been some forra of Powisartmh ?

THE NINE WITCHES OF GLOUCESTER

293

his biography. For that has its parallels in Irish legend : compare,
for instance, the story of Corpri Muse in the Book of the Dun Cow,
fo. 54% see also ' Celtic Heathendom,' pp. 308, 309, and ' The Welsh
People ', pp. 36, 37,

There remains the question why the Goidelic warrior women
were described in Welsh as ffmddomd. It is unfortunate here
that the etymology of that word eludes my search ; but the texts
leave no doubt that in them gimddon was capable of bearing the
signification of witch or sorceress. They do more, they associate the
Qwiddmwd with prophecy and prediction in a way which allows one
to suppose that they were regarded as the authors of them. The
room for doubt is narrowed by the way one of them recognized
Feredur at his first encounter with them : the recognition was so un-
expected as to take Peredur himself by surprise, which he betrayed
in his question, * Why, hag, knowest thou that I am Peredur ? ' The
Witches have already been compared to ScAthach, the amazon in
Britain who taught Ciichulainn feats of anus. Her dominion
bordered on that of one whose queen, named Aiffe, was also a great
sorceress ; but with regard to Scdthach we are told explicitly that
she was a fdith^ that is to say a mtes or prophetess. Wlien there-
fore Cuchulainn had learnt all that she could teach him, at Ms
departure she foretold in a poem — a very obscure one — the events
of his life, including the exploits and hardships described
in the Tain B6 Cmlnge^ the great epic story of Irish Hterature.
That she is represented doing by means of the process known to
Irish magic as Imbas ForosnUj to which recourse was also had by
Fedelm, a prophetess consulted by Queen Maive when setting out
on the Tdin. Fedelm likewise had learnt her business of poet and
prophet in Britain : in Ireland the Tmhm Forosna was eventually
forbidden by St. Patrick because it involved an oflFering to the
gods of the pagans. In the light of the Irish parallel of Sc^thach,
one can hardly be wrong in treating the Witches of Gloucester as
Goidelic sorceresses who were regarded as enjoying the gift of
prophecy and prediction,^

' As to ScMhach see the Book of the Don Cow, fo. 125^ 126^' Tochmarc
Emire (ed. K, Meyer), Heme CeUiqtie, si. 448, 449, 452, 453, 457 j O'Cyrry's
Manners and Customs^ ii. 369-71, As to Fedelm see the Book of the Bun Cow,
fo. 65^' ; D*Arbois de Jobainville, Eev. CeU,^ xxviii. 155-9. For a description of
Imbas Forosna and its abolition, see Cormac's Olossaiy, b. y. Imbas.

WHO WEEE THE DORIANS?

By WILLIAM RIDGEWAY, M,A, RB,A, Hon. D.Lrrr.

DmNEY Professor of Archaeology and Brereton Reader in
Classics in the UNrvERsrrY op Cambridge

In my Early Age of Greece, voL i, pp. 287 sqq., I argued that
the lower part of the Balkan Penmsula, which we know as Greece,
was inhabited fmm the Neolithic period by the same melanochrous
race, which stiU forms by far the largest element in its population,
that these people were the authors of the great civilization of the
Bronze Age, commonly termed * Mycenean ', or by some ' Minoan ';
1 further urged that this people never spoke any other than an
Indo-European tongue, and that it is their language which we find
in the various dialects of both ancient and modern Greece ; I further
argued that a body of tall fair-haired immigrants had come into
Greece from the Danubian and Alpine regions somewhere about
1500 B.C., and that these people, known to us as Achaeans, were part
of the great fair-haired race of upper Europe termed by the ancients
the Keltoi, and now commonly described as Teutonic. This people
brought with them the use of iron, they burned their dead instead
of burying them as did the aborigines, they had garments of a
diffei-ent kind, which they fastened with brooches, and they brought
with them a peculiar form of ornament, which is commonly termed
geometric or dipylon. I have also pointed out elsewhere that they
differed essentially in their social institutions and religion from the
Pelasgians, whom they conquered. There can be no doubt that the
aboriginal race traced descent through females, as was certainly the
case at Athens, in Aicadia and in ancient Argolis, which can be proved
from the legends themselves.

On the other hand, the Homeric Achaeans are strictly monan-
drous, the fideUty of Penelope having become proverbial through
the ages, whilst their wives expected similar constancy on the part
of their husbands, as is shown by the story of the jealousy of the
mother of Phoenix, and the part played by that hero in esiK)using his
mother's cause against his father.

296

WHO WERE THE DORIANS?

I likewise pointed out that all the lUyrioThracian tribes of the
upper Balkan belonged to the same melanochrous race as that of
Greece^ speaking likewise an Indo-European language.' But those
tribes had been conquered in many instances by Keltoi from the
Alpine regions, or else driven out completely. Thus the Getae and
Trausi were certainly not ethnologically Thracians, though so
termed geographically, for they were the * red Thracians \ and w©
know from Herodotus that they differed essentially in all their
customs from the indigenous Thracians whom they had conquered.
Moreover, it appeared that the ruling families in most of the
Thracian tribes were of this other stock, for they had gods different
from those of their subjects. Such then is the ethnical condition of
the Balkan Peninsula at the dawn of history.

The Achaeans remained noasters of Thessaly until they were
driven out or subdued by the Thessalians, an Illyrian tribe, who
crossed the Findus in 1124 b.c*, whilst the Achaeans continued
dominant in Argolis and Laconia until the Dorian invasion some
twenty years later in 1104 Ra

It was universally assumed, until the appearance of my Early
Age of Greece^ that it was to the Dorians we ought to ascribe the new
form of ornament known as the geometric, to which I have just
alluded. But I was able to show that it had already had a com-
plete grip upon Peloponnesus before the Dorians had ever planted
foot in that region.

It has been generally assumed by my reviewers that the
Dorians are simply another wave of the same stock as the Achaeans,
and they have taken for granted that I hold that view. This paper
is an endeavour to make clear who the Dorians were, and to show
that they are in nj> wise to be regarded as belonging to the same i
Keltic stock as the Achaeans. The facts, as I read them, point
entirely in a different direction, and I hope to show tliat the
Dorians, like the Thessali, were simply an Illyrian tribe. • Space
renders it impossible for me to develop at length all the arguments
which can be marshalled to support this doctrine, and I wUl there-
fore simply try to show that the Dorians differed essentially from
the Achaeans in {1) their social system and law of succession, (2) their
physical characteristics, (3) method of wearing their hair, (4) shaving
the upper lip, (5) burial of the dead, and (6) in their language, or rather
dialect, whereas in all these respects they agree with the lllyriana

* Op. cit., voL i, pp. 342 sqq.

WHO WERE THE DORIANS?

297

According to Herodotus/ the first home of the Dorians was in Phthiotis.
From thence they migrated (probably driven out by the Achaeans)
under the mythical leaderahip of Donis, son of Hellen^ to the tract at
the base of Ossa and Olympus, called Histiaeotis. Forced to retire
from that region by the Cadmeans (who were almost certainly the
Thracian tribe known as Phlegj^ans in Homer ^) * they settled under
the name of Macedni in Findus. Hence they once more removed
and came to Dryopis, and from Dryopis, having entered the
Peloponnesus in this way, they became known as Dorians/ Else-
where Herodotus ^ speaks of the Lacedaemonians, Corinthians, Sicyo-
nians, Epidaurians, Troezenians, and Hermionians as * Dorians and
Macednians ', except those of Hermione, and as having emigrated
last from Erineus, Pindus, and Dryopis. The Hermionians were
Dryopians * whom Heracles and the Malians had driven out of the
land now called Dorus *. The region known as Doris or the Doric
Tetrapolis in classical times was the district lying between Thessaly,
Phocis, and Acarnania. But not all the Dorians of Histiaeotis had
migrated to Pindus, for ancient tradition states that some of them
had sought new homes in Crete, and it is those settlers who
are mentioned in the only place where the Dorian name occurs in
Homer, the famous lines in the Odyssey* which give that invaluable
description of the ethnology of Crete,

Herodotus thus held that the Dorians were Macedni or Macedonian
in stock. But as the aboriginal Macedonians were closely identified
in speech, dress, and method of wearing their hair with the Illyrian
tribes by the ancients themselves, this statement of Herodotus
makes a prima facie caae for regarding the Dorians as Illyrians.

The Dorians who invaded the Peloponnesus were under the leader-
ship of the Heraclidae, who certainly regarded themselves as dififer-
ing in race fi-oni their subjects. Thus, w^hen Cleomenes,'* the Spartan
king, occupied Athens in 609 b,c., on his attempting to enter the
temple of Athena on the Acropolis the priestess forbade him, on the
ground that it was not lawful for a Dorian to do so. Thereupon the
king repMed, ' I am not a Dorian, but an Achaean/ Clearly then
the Spartans themselves knew that there was a strong racial dis-
tinction between Dorians and Achaeans. I have pointed out that
the Illyrian and Thracian tribes were in many cases imder the
chieftainship of men of the Keltic stock from the Danubian region.

i. 56,

• n. xui, 301. ' viii. 48.

' Herod, v. 72.

* xix. 177.

298

WHO WERE THE DORIANS?

and as the Achaean kings had intermarried freely ^ith the ancient
royal houses of Peloponnesus to which Heracles belonged, e.g.
Pelops, Atreus, Agamemnon and Menelaus, we are not surprised
that the Heraclidae laid claim to Achaean descent, and that the
royal family of Sparta in classical times called themselves Achaeans,
and not Dorians.

Social System. — The Achaeans of Homer are represented
as strictly monandrous. As might be naturally expected, in such
social conditions there is no trace of anything but male descent,
amongst them. Each chief has succeeded his father and expects to
be succeeded by his own son, as is clear in the case of Peleus and
Achilles, Laertes, Odysseus, and Telemachus, Atreus, Agamemnon
and Orestes. Thus the doctrine of paternal descent as well as
of monandry is as strongly marked amongst them as it was
amongst the Germanic tribes of a later period. In Homer we have
certainly cases of descent through women, but they are found
amongst the Trojans and the Lycians, who of course were not
Achaeans, the Lycians continuing to be typical examples of descent
reckoned through females down to late classical times.

It can be shown that all the melanochrous peoples of Europe at
the dawn of liistory were addicted to polyandry, and had descent
through women as a natural corollary of their loose social system.
Such was the case amongst the dark-haired inhabitants of Britain *
and Ireland -^ in Spain, in the Balearic Islands, amongst the
ligurians^^ of France and Italy, and amongst the Illyrian and
Thracian tribes other than the Getae and the Trausi, who, as we
saw, were really Kelts.

We learn from Herodotus* that the lUyrian tribe of the Veneti
had the practice of selling the girls of each \dllage by auction for
wedlock, when the handsome girls were made to supply dowers to
obtain husbands for their less comely companions. Prom this we
may infer that the girls were allowed complete licence before
marriage, and that a man had to buy the right of having one of
them assigned to him for his exclusive use. This inference is com-

» Caesar, 5.G. v. 14.

* Kidgeway, ' The Date of the first shaping of the Cuchulaion Sagm' p. 2
{Proceedings of BrHisli Academy, voL ii) ; Strabo, 167. 19(Didot),

' Ridgeway^ * Who were the Romana ? ' (iVoc. of British Academy^ voL ui).

* L 196,

WHO WERE THE DORIANS?

299

pletely confirmed by the further statement of Herodotus ^ that the
Thracians permitted the utmost licence to their girls before they
were purchased for marriage* Moreover, the Ulyrians seem to have
had female chieftains, for it is significant that the most powerful
Dlyrian sovereign of whom we read was Teuta, the Illyrian queen
who in 230 ac, put to death the envoys of Rome3 Probably
Aristotle had in his mind the Illyrians as well as the Thnicians
and Dorians, when he said tliat * the warlike races, with the exception
of the Kelts, were under the control of women'. The fact that the
Illyrians had female monarchs in the third century ac, combined
with the statement that in the fifth century b.c. the Veneti held that
all the girls of the community were common property, and with the
fact that polyandry was universal amongst their closely connected
Tlaracian tribes, leaves little doubt that polyandry also existed
amongst them as it did amongst the Agathyrsi, who dwelt in what is
now Transylvania, and who were closely akin to both Thracians and
lUyrians. These men had even their wives in common * in order
that they might be all brothers and as members of one family
might neither envy nor hate one another'. They had thus antici-
pated Plato s famous doctrine.^

The legendary histoiy of the Illyrians corroborates this con-
clusion in a notable fashion. Although there was every temptation
to furnish each Illyrian tribe with an eponymous hero, many of
them had a heroine at the head of their pedigree. The sons of
Ulyrius were said to be Encheleus, Autarieus, Dardanus, Maidus,
Taulantus, Perrhaebus, and his daughter were Pai'tho, Daortho,
Dasaro, and others from whom are sprung the Taulantii, the
Perrhaebians, and the Enchelees, the Autarieis, tlie Dardanii, the
Parthenii, the Dasaretii, and the Darsii.^ It is noteworthy that the
Perrhaebians, who were amongst the most ancient tribes of Thessaly^
are regarded as Illyrian, whilst the names of the heroines all end
in -co, the regular termination in Doric female names, such as Gorge,
Lampito, &c.

Let us now turn to the Dorians of the classical period, and let
us examine the legendary and historical evidence bearing on the
relation of the sexes among the Dorian aristocracies of Laconia and
Argolis,

Y. &

* Flmy, H.N. xriv. 24.

* Appian, Illyr. 2.

Herod, iv. 49, 100, 104.

800

WHO WERE THE DOKIANS?

Fortunately both Xenophon,' and Polybius,^^ have left us
statements about the meaning of which there can be no doubt.
According to the former, conjugal fidelity was practically unknown
at Sparta ; this he ascribed to the legislative enactments of LycurguSj
who directed all his attention to produce and rear a vigorous brood
of citizens. The lawgiver is represented as regulating the age of
marriage for the sexes, and as he saw that when old men had young
wives they exercised a special surveillance over them, he ordered, on
the contrary, that an old man should bring in some man pre-eminent
in physique and courage, and should get him to procreate children
for him. Again, if a man did not care to cohabit with his mfe, but
should desire a child, he ordered that in case his eye had lighted
on a fine woman, who had borne fine children, he was to get her
husband's permission and have children by her. Xenophon adds
that ' the women are ready to be mistresses of two houses, and the
men to give a share in their children to their brothers, who partici*
pate in the family and power but do not make any claim to the
property '.

The obscure sentence which I have paraphrased by the words
* give a share in their children to their brothers ' is rendered
perfectly lucid by Polybius^ who states that it was * customary with
the Lacedaemonians for three or four men, or sometimes more, if
they were brothers, to have one wife, and it was esteemed right for
a man in case he had begotten a sufficient number of sons, to hand
over his wife to one of his friends '.

The specific statements of Xenophon and Polybius are fully borne
out by the famous passage in which Aristotle ^ criticizes the Spartan
constitution : * Again the licence of the women at Lacedaemon is
equally fatal to the spirit of the polity and to the happiness of the
state. For as husband and wife are constituent elements of a
household, it is right to regard a State also as divided nearly
equally into the male and female population, and accordingly in

* Rep* Lac. i. 9 : at tc yap yiTatice? Sirrov? oikovs ^ovkovriu icarfx^iVt ot r€ avSp€^
dScX^ov^s rots waari 7rpo<rXafipdv€tVf ot rov filv yivdv^ koX ttjs Bvvdfuuj/^ koivwvovo'i riuv Sc

^ Mai, Scripiorum Veierum Nma Cdkctio e Vaticanis codidhus ediia, Tom. U,

p. 384. 11 a pa ^ty ovv Toli Aat(€^atfwvtoi^ *fat wdrptjov tjf Mat irvvrjOt^ rpti^ avSpai €)(€iv
yi^vaiica nai rirrapa^' tcStc §€ ttal irXcmvs dScA^i?* oinaSf koX ritiva rovTiay ttyat Koiyd* Kai
ytVYTJ^ravra TratSa? t^avov?, iK^vvat ymfama rvyt rmy ^jiCktuv icoAok hqX ot^'tjOis (Polybll
Exctfpta, lib. xiii, fragm. vi). =* Pol. ii. 9.

WHO WERE THE DORIANS?

301

any polity where the condition of the women is unsatisfactory, one
half of the State niubt be regarded as destitute of legislative regula-
tions. And this is actually the case at Lacedaemon. For the
legislator, in his desire to impart a character of hardness to the
State as a whole, although true to his principle as regards the men,
has been guilty of serious oversights in his treatment of the women,
as their life is one of unrestrained and indiscriminate hcence and
luxuiy/

We have thus irrefragable evidence of not only general
polyandry but also of the more limited Tibetan form, wherein
several brothel's have the family property and one wife in common.
Nor can it be urged that this phase of society had sprung up in
Sparta at a comparatively recent date. For according to Aristotle
Lycurgus *made an effort to reduce the women into conformity
with the laws, but they resisted so stoutly that he abandoned the
attempt \ This is, of course, at variance with the statement of
Xenophon, who ascribes to Lycurgus the great laxity of Spartan
domestic life. Both stories, however, assume that polyandry had
existed in Sparta from a very early stage of her history.

McLennan * cited that story which represents Lycm^gus as
declining, on purpose to set an example to his countr3rmen, to marry
his brother's widow and cut out from the succession his brother s
son, as indicating the transition from female to male succession in
the royal Dorian house. Though this story cannot be regarded as
proving McLennan's contention, yet we have no difficulty in finding
an undoubted case of female descent in another great Dorian family.
But if descent had once been reckoned through women, it is a sure
evidence that laxity of sexual relations had once been the regular
order of things.

Since the postnuptial unchastity of the Spartan women is
so clearly proved, we have no reason to doubt that their prenuptiai
morahty was as low as it is painted by Euripides in Ins Andromache:-

Nor is it only in Sparta that there is proof that polyandry
and female kinship once existed among the Dorians. Corinth like-
wise supplies some evidence of importance. In that city kings had
been superseded in the course of time by annual magistrates named
Prytanes. These were chosen from the clan of the Bacchiadae,
* who intermarried only among themselves and held the manage-
ment of affairs. Now it happened that Amphion, one of these, had

* Studies in Ancient History, p. 232, • 559-95 aqq.

302

WHO WERE THE DORIANS?

a daughter named Labda, who was lame, and whom therefore none
of the Bacehiadae would consent to marry ; so she was taken to
wife by Aetion, son of Echecrates, a man of the Deme of Petra,
who was, however, by descent of the race of the Lapithae^ and of
the house of Caeneus/" Labda bore a son, whereupon the
Bacehiadae sent ten of their number to destroy him, but his
mother hid him in a chest (kypsele)^ whence he obtained the name
of Cypselus. Unless the succession of the Bacehiadae had been
through women there would have been no reason for their careful
endogamy. On the contrary they would have been able to
strengthen their power by giving their daughters in marriage to
leading men among the other citizens.

Again at Argos and Epidaurus in addition to the three familiar
Dorian tribes — Hylleis, Dymanes, and PamphyH — there was a
fourth named Hymathia, which may perhaps have been named
after Hymetho, the daughter of Temenus, the first Dorian king of
Ai^olis. 'This Hymetho was married to Deiphontes, whom
Temenus had openly employed as his general in his battles instead
of his own sons, and he took liis advice in everything ; and as he
had previously made him his son-inlaw and loved his daughter
Hyrnetho best of all his children, he was suspected of trying to
divert the kingdom to her and Deiphontes. Therefore his sons
plotted against him, and Cisus, the eldest of them, moimted the
throne \^

Here we have apparently not only a tribe named after a woman,
as amongst the lllyrians, but a clear indication of a time when
female succession was the rule amongst the Dorians. Another
legend indicates that the feeling of the nation was in favour of the
daughter Hyrnetho and her husband, for we are told by Pausanias ^
that Deiphontes and the Argives took possession of Epidaurus. The
latter had separated from the rest of the Argives after the death
of Temenus, because Deiphontes and Hyrnetho hated the sons of
Temenus, and their army was more attached to them than to Cisus
and his brothers. Thus in both Argolis and Laconia there is no
lack of evidence that the Dorians at the time of their conquest had
still the rule of female kinship.

In face of this evidence it cannot any longer be maintained
that the Homeric picture of society reflects the social life of the
Dorians at any period of their history. The facts demonstrate that

' Herod, v. 92. ' Paus. ii. 19^ 1. iL 26, 1. » iL 26, L

WHO WERE THE DORIANS?

303

the Dorian invaders of Peloponnesus were no less polyandrous than
the aboriginal Pelasgian population of that area, ■

That two-fifths of the land in Laconia was in the hands
of women, so much commented on by both ancient and modern
writers, was probably due to the ancient custom of female
succession, and an hereditary readiness on the part of the Dorians
to leave their land to their daughters, even at a time when descent
was now reckoned thTOUgli males. -^

C. O* Milller * has pointed out that amongst the Thessalians, <
who, as we have seen, were certainly Illyrian in origin, the women,
as amongst the Dorians, were addressed by the title of ' ladies '
(heawotpai^si foiTii of address * uncommon in Greece and expressive
of the estimation in which they were held '.

Again, the same writer has pointed out that the love of the m^e
sex (that usage peculiar to the Dorians) was also common amongst
the Thessalians, and that the boy-favourites were called by the same
name (dtrai) as at Sparta- This similarity of nomenclature is very
remarkable, and seems to point to a very close relationship between
the Dorians and lUyrians,

Physical Characteristics, — The Achaeans of Homer were
large men with fair hair. But it is almost beyond doubt that the
Dorians were neither remarkably tall nor xanthochrous. For if
such had been the case, as it was with the Thebans, both these
points would certainly have been remarked in some of the various
passages in classical authors which refer to them, especially in view
of the frequent reference to Spartan women, their habits and their
costume. Yet the Pseudo-Dicaearchus gives us a full account of the
physical characteristics of the Thebans, and from his description it is
reasonable to infer that the fair hair and tall stature of that people
were quite exceptional in Greece, a statement hardly likely to have
been made had the same features marked the Spartans.

There are several well-known passages in Greek authors, which
afford every opportunity for reference to the colour of the hair of
Spartan men and women, if there had been anything unusual or
striking in it. For example, though Herodotus ^ relates how the
Persian scout observed the Spartans combing their hair on the
eve of the final struggle at Thermopylae, it is strange that with all
his love of detail the historian should not have alluded to its colour

* The BorianSj vol i, p. 3.

Ibid.

vii, 208.

304

WHO WERE THE DORIANS?

if it had differed in shade from that of other Greeks* Again, in the
Lyststrata of Aristophanes * there is every opportunity for allusion
to the colour of the hair of Spartan ladies, had it differed in any way
from that of the Athenians. For when Lampito, the Spartan lady
delegate to the Peace Conference, arrives at Athens, though her
Athenian hostess Lysistrata makes some very minute observations
on her personal appearance, there is not the slightest reference to
the colour of her hair. Now, as at that time it was the fashion at
Athens, as we know from the same comedy/ for ladies to dye their
hair yellow and to wear * Cimbrian ' garments in imitation of the
Galhc women of the upper Balkan, Lysistrata would certainly have
alluded to the fasliionable colour of Lampito's hair, had she been
blonde. Though the argument e silentio is often dangerous, yet when
there are so many passages where reference to the colour of the
Dorians' hair would have been made had it been unusual, we may
conclude with very great probabiUty that it in no wise differed from
that of the ordinary nielanoclirous population of the Balkan peninsula.

Fashion of wearing the Hair, — It will be naturally said
that the passage from Herodotus just cited proves that the Spartans
wore their hair long just as did the long-haired {KapT} Kofiompre*;)
Achaeans of Homer, and that accordingly in this respect they closely
resemble the latter people. But it must not be assumed that because
the Spartans in the fifth century b. c. wore their hair long, as did the
Homeric Achaeans, they are thereby to be identified as belonging to
the same ethnic group, for Herodotus ^ expressly tells us that the
Spartans had only adopted the practice of wearing their hair long
after the overthrow of the Argives in the struggle for Thyrea, having
up to that time cut their hair. But whilst the Celts of the Danubian
and Alpine regions wore their fair hair unshorn as did the Achaeans,
on the other hand the Illyriao and Thracian tribes cut their hair.*
Thus, then, the Dorians agree in complexion and custom of
cutting the hair with the Blyrio-Thracian tribes which bordered on
Thessaly,

* Xy*. 78*83 : AY2. olov to koXXo^^ ykvKvrdTTit trov <j>a(y trau

<iis S' €V)(po€L^f fa>s S( «r<^pty^ TO <ritifbd crov,

K^v ravpov aty;(ot9. A AM. p.dXa y olHi vol To* <nta*

YVfi,va8^fMiL yap teal wort irifyav oAAofuiu

AY2» telS ^ KoAoK TO XP^H'* *X*'^ ''**'*' TtT^iClJV*

' I^s. 43-5. » i. 82.

* Ridgeway, Earltf Age of Greece, vol. i, p. 344 ; c£ Theocr. xiv. 46.

WHO WERE THE DORIANS?

305

Shaving the Upper Lip* — The Spartans shaved their upper
lip, as we know from Aristotle/ wliilst Plutai^ch - tells us that
the ephors on coming into office enjoined on all men ' to shave the
moustache ' and to ' obey the laws \ On the other hand, the Achaeans
of Homer seem not to have shaved the upper lip at all, wliilst from
the earliest date at which the Celts from the Danubian region come
within our ken, they are especially distinguished by weaiing the
moustache, as is well exemplified in the famous statue of tlie
Dying Gaul.

Disposal of the Dead. — I have elsewhere pointed out ■' that
the Dorians did not burn their dead, as did the Homeric Achaeans,
but inhumed them, as was the practice of the indigenous Illyrians
and Thracians.* On the other hand, the Celto-Umbrian tribes of
central Europe and upper Italy always cremated their dead, as did
the Achaeans,

Plutarch ^' tells us that Lycurgus made excellent regulations for
funerals. In the fii'st place, in order to kill superstition, he raised
no objection to burying the dead in the city and having their
monuments near the temples, thus habituating the youths to such
sights, that they might not be perturbed by or shrink from death as
though it defiled those who touch a dead body or pass athwart
graves. * In the second place he forbade anything being buried
along with the dead, but they used to bury the body wrapt in a
purple cloak and olive leaves,' Elsewhere '' Plutarch says that when
an ordinary Spartan died in a foreign land it was the custom to
celebrate his funeral rites there and leave his body behind, but the
bodies of kings were brought home. Thus the Spartiates, who were
with Agesilaus when he died far from home, melted wax all over the
corpse, since honey failed them, and brought it back to Sparta. Nor
can it be said that it was only the Dorians of Sparta who practised
inhumation. The Megarians interred their dead, but they seemed to
have laid tlie body to face east, though according to Heraeus of
Megara the Megarians also buried the corpse to face the west', and
moreover shared with Salamis the custom of laying three or four in
one tomb. The double usage m orientation at Megara was probably

' Be Sera Num^ VincL 550 (Eei&ke) ^17 T/>c<^f tv /njerroica koI w€i0€<T$at TOi« Mo/totf.

* Earltf Age of Greece^ voL i, pp. 490-1.

' Ibid, p, 495, ^ L^c 27. • Ages. 40,

306

WHO WERE THE DORIANS?

due to the mixed aboriginal and Dorian population. As the dead
lay towards the west in Attica, Salamis, in some cases at Megara,
and in the majority of the interments in the Shaft graves at
Mycenae, it may be inferred that this was the characteristic
orientation of the autochthonous race.

The Sicyonians buried their dead * in the ground ', though of
com*se these might be the aboriginal race, since they liad been able
to overpower their Dorian masters.

Dr. A. J, Evans * has lately found at Cnossus, * about a mile to
the north (of the Palace) on the way to the headland where a Royal
tomb had already been excavated, a series of *^ Cyclopean " blocks *
which proved to have been removed from their original context.
Immediately below them were two beehive tombs cut out of the soft
rock. In their form and certain features of their contents they
represent * the old Minoan tradition \ but they belong * to a period
about 8CM) b. c, when the Dorian settlement of a large part of the
island was already an accomplished fact. The swords here were of
the mainland type, iron succeeding the earlier bronze, and cineraiy
urns had replaced the earlier corpse burial j but the variety and
invention displayed in the objects found, the continuity of many of
the decorative motives, as well as the appearance of the characteristic
** sturup vase ", pointed to a distinct survival of the old indigenous
element. In one tomb there were nearly a hundred vessels, and
among them the more important cinerary urns presented quite a new
and very elaborate style of Geometrical design.'

In view of the facts just cited above, Dr. Evans's idea that these
cinerary urns are those of Dorians must be rejected. On the other
hand they exactly fit the Homeric period, when the Achaeans, who
preceded the Dorians on the mainland, always cremated their dead.
Moreover, as the Odyssey represents the Achaeans in possession of
Crete at the time of the Trojan war, whilst there was no large influx
of Dorians from the mainland until long after they had conquered
the Achaeans in Argolis and Laconia, the continuity of the older style
of decoration as well as of the ' stirrup cup vases ' can be much better
explained, if the new tomb belonged to the Achaean lords of Cnossus,
who had overthrown the Minoan dynasty.

Dialect. — There are ceitain labialized forms in Homer and in
later Greek which I have compared^ with the similar labialized

» Times, July 15, 1907. * Op. cU., 078-4.

WHO WERE THE DORIANS?

307

forms in use amongst the Celts and their close kinsmen, the
Umbrian-Sabellian peoples of Italy. For instance, wia-vpt^ for
rirrapt^ in Homer and the form Itttto^ common to all the later
Greek dialects^ which had certainly replaced an older form lkko%
whose existence is proved by the ancient lexicographers. The best
modern philologists are agreed that the form Imro^ has come into
Greece, and I have compared its labialized form with the Gallic
Epona (the horse-goddess) and the modern Welsh eh (hoi'se). When
the Achaeans were driven out of what was later known as Thessaly
by the invading Ill3rrian tribe of Thessali nearly at the same time
as the Dorian invasion of Peloponnesus some of them and their
Achaeanized subjects settled in Boeotia. Doubtless it was to them,
and possibly to the Phlegyane, who had previously captured Thebes
and settled there, that Thebes owed her famous tall, fair-haired
inhabitants. It is interesting to find in this very area some
unmistakably labiaUzed forms. Thus the Boeotians said Trerrapc?
instead of rerrapes, they called a woman ^avd^ not yi/inj, as did the
Athenians and the lonians, or yi^i/a, as did the Dorians, Again, the
Boeotians called the locust Tropwail^, as did also their kinsmen who
had settled in the Aeolid in north-west Asia Minor, when others of
their number had found new homes in Boeotia, %vhile it is certain
that the indigenous population of Greecej such as the people of
Mount Oeta, used the unlabialized form Kopvo^^

The Boeotian form ircrrapc^ = the Gallic petor found in petw-
ritum, * a four-wheeler,' and the Umbrian peiur^ whilst ^aud
corresponds to the Irish ban, * woman/ It is now clear that as
the Dorians do not use any of the labialized forms peculiar to
Boeotia and the Aeolid whither the Achaeans from Thessaly had
migrated, they cannot have been part of the Achaean stock which
had entered Greece centuries before the Dorian migration. On the
other hand there is no evidence that the Illyrians had any tendency
to labialisnij for the scanty available data point rather to their being
distinctly a K folk, as were their close neighbours the Ligurians,
who formed the oldest stratum of population over a great part of
Italy, just as the Illyrians did over a large part of the sister
peninsula.

Though the Spartans thus differed essentially in their phonetics
from the Achaeans of Homer and from the fair-haired folk who
formed a leading factor in the population of a large part of Boeotia

* Strabo, 524, 28 (Didot).
X 2

ao8

WHO WERE THE DORIANS?

in the classical period, they undoubtedly agreed with the lUyrians
in at least one characteristic.

The Spartans, as is well known, used a where all other Greeks
UBed d, e. g. crtos - dto^. Now the Macedonian name for Silenoi
was AavdSai, whilst the lUyrian term for the same was ScvctSat.
Again, there was an lUyrian tribe, the next neighbour to the tribes of
Macedonia proper, called Aao-apcVtot who were also called Sesarethii
{ttaapr^Oioi)} The occurrence of the parallel forms Datmdui and
Seuadai forbids any rash emendation of either form into the other,
as well as the explicit statement of Strabo that there was a double
form of the name. It is plain that d cannot come from s nor s from
d^ whilst it is equally certain that both can come from a common
DH, For in Macedonian the medials regularly represent the
original aspirates, e. g. BcXtiriro? ~ ^IXnrwo^, Bep€viKo^, B€p€viKr}
{B€ppiK7}} = <f>€p€PtKo$, <p€p€viKri. Thus a conunou form ^auciSag* would
give Macedonian AauaSag and Illyrian l€vdSas, and a common form
Baa-apTjTioL^ woidd give both Dasaretii and Se^arethii^ the former being
almost certainly the form used by the Macedonians who bordered
on this tribe, the latter by the lUyrians themselves.

Thus we have fully proved for Illyrian the assibilation of original
DH, which characterizes Doric alone amongst the various Greek
dialects. The fact that both Dorians and Illyi^iaus had in common
this feature not found in the dialects of their neighbours points
unmistakably to their very close relationship.

We have already seen above that the names of the heroines
fix)m whom many Illyrian tribes traced their descent all end in
-0), which is also the regular termination of female names in
Doric.

There are other pieces of evidence which cannot here be
adduced owing to the conditions of space, but I venture to think that
enough has been said to show that the Dorians differed essentially
in race from the Achaeans of Homer, whilst they so closely resemble
in their social habits, their physique, method of wearing the hair, in
the disposal of their dead, and in theii* dialectic forms, the lUyrians,
that they must bo regarded, like the Thessali, as an Illyrian tribe.

Strata, 271, 23 (Didot).

ON THE ORIGIN OF THE CLASSIFICATORY
SYSTEM OF RELATIONSHIPS

By W, H. R KIVEES,
Fellow of St, Johns College^ Cambridge

Lewis Morgan is the only modern writer who has attempted to
formulate a complete scheme of the evolution of the human family,
a scheme based almost entirely on a study of the classificatory system
of relationships of which he was the discoverer. According to thiB
scheme human society has advanced from a state of complete pro-
miscuity to one characterized by monogamy by a gradual evolution,
the three chief stages of which Morgan called the consanguine, the
Punaluan, and the monogamian families. In recent years the scheme
has encountered much opposition, especially fiom Starcke,* Wester-
marck,^ Crawley,^ Andrew Lang/ and N. W. Thomas/' the last
calling Morgan s whole structure a house of cards, and it may
perhaps be said that the prevaiUng tendency in anthropology'^ ia
against any scheme which would derive human society from a state
of promiscuity, whether complete or of that modified form to which
the term group-marriage is usually applied.

The opponents of Moi'gan have made no attempt to distinguish
between different parts of his scheme, but having shown that certain
of its features are unsatisfactory, they have condemned the whole.
The elaborate scheme of Morgan can be divided into two distinct
parts, one dealing with the existence of the consanguine family and
the evolution from this of the Punaluan family, while the other part
deals with the existence of this latter form of the family itself. It
will be my object in this paper to point out a radical defect in the
fii'st part of Moi^gan's scheme, and then to endeavour to restate the

^ The Primitim Family, London, 1889*

^ History of Human Mamage, 3rd ed*^ 1901,

* The Mystic Hose, London, 1902.

* Social Originsj London, 1903, p. 90.

^ Kimhip Oryanisalions and Group Marnage in AustraUa, Cambridge, 1906.
' The chief ©xoeption among those who have written on this subject in recent
yeai^ is Kohler ; aee Zur Urgeschklite cCer Ehe^ Stuttgart, 1897.

310

ON THE ORIGIN OF THE

second part of Iiis scheme in accordance with the knowledge which
has accumulated since his time.

The existence of both the consanguine and Punaluan families was
deduced by Morgan from the characters of the classiticatory system
of relationships. This system is found throughout the whole of
North America, and probably exists also in the South. It is universal
throughout the Pacific^ — in Polynesia, Melanesia, New Guinea, and
Australia. It is found in India, and some typical examples have
been reported from Africa, over which continent it is probably very
widely spread. Vestiges of it are found in other parts of the world,
and it is probable that relationships have been expressed in this
way by all the races of the world in the early stages of their
development. The most important feature of the system is that
large groups of people who, according to our ideas, are related in very
dififerent ways and in very different degrees are all ranged in the same
category. The same name is given to a distant cousin once removed,
for example, as is given to the father. On the other hand, relatives
who are given the same name by most civilized people are in the
classificatory system often rigorously distinguished. In this paper
I propose to consider how far there is reason to believe that this
system had its origin in the organization of eaily society, and
especially in the early modes of relationship between men and
women. In the first part of the paper I shall deal with the
evidence provided by the system for the existence of Morgan's
consanguine family, and in the second part shall consider the origin
of the system in a condition of group-marri^e.

The Nature of Mofyans Makiyan SystenL

Morgan's belief in the existence of the consanguine family,
which corresponds to what is often called the undivided commune,
Wiis based entirely on the view that the variety of the classificatory
system which he called Malayan * was the earliest form of the
system. If it can be shown that the Malayan form represents a
late stage in the development of the system, the whole evidence for
the consanguine famOy falls to the ground so fai* iis it is provided by

^ The actual examples on which Morgan based his Malayan system were from
Polynesia, the name Malayan being chosen by him because he regarded the Poly-
nesians as a branch of the Malayan family (Afwient Sockt^^ p. 403), In spite of
much recent w^ork on the Malays we are still almost wholly in the dark as to the
kind of kinship system found among the different branches of that people.

CLASSIFICATORY SYSTEM OF RELATIONSHIPS 311

the classificatory system, and Morgan himself acknowledged ^ that
Ills hypothesis of the consanguine family rested principally, if not
wholly, on this foundation.

Morgan supposed that the Polynesian societies which posse.ssed
the Malayan system were in a pristine state of culture, and he
believed that their system of relationships revealed a coiTesponding
primitive state of the evolution of the human family. We now know
that Polynesian society is relatively highly developed, and it may
perhaps be held to be superfluous to show that their kinship system,
instead of being archaic as Morgan supposed, is a late product of
change* I have been unable to find, however, that any student of
the subject, whether supporter or opponent of Morgan, has refused
to accept the Malayan foim as primitive, and since the belief in its
priniitiveness is at the bottom of many of the difficulties in con-
nexion with this subject, the evidence in favour of the lateness of
the system may be given.

The special chai*acteristic of the Malayan or Polynesian system
is the small number of terms and the corresponding wide connota-
tion of each. Tlie same terms are used to denote relationships for
which many different terms are found in most forms of the classifica-
tory system ; thus, excluding diflferences dependent on age and sex,
all the relatives of a speaker of the same generation as himself are
addressed by the same name. The distinctions between fathers
brother and mother s brother and between father's sister and
mother's sister which are usual in the classificatory system are not
present, and there is a corresponding absence of distinctive names
for their children. Morgan supposed that we had in this system the
survival of a state of society in which all the members of a group
corresponding to the brothers and sisters of a later stage intermarried
indiscriminately, the consanguine family which he advanced as tlie
earliest stage of human society,

I hope to show that this wide connotation of relationship terms
is late, and not primitive, by pointing out that elsewhere w^e fmd
examples where classificatory systems are undergoing changes which
are modifying them in the direction of the Hawaian form. My
attention was directed to this problem by a study of the relationship
systems of Torres Straits. We have in these islands two peoples in
different conditions of social organization. In both there is patri-
lineal descent, with fairly definite evidence in one case at least that
' Ancient Socieiif, pp, 385, aSS, 402.

312

ON THE ORIGIN OF THE

the people have emerged from a pre\ious condition of mother-right,
and the high degi-ee of development of the idea of property would
seem to indicate that their social condition is far from being of a
primitive kind. On examining the social organization of the two
communities we find additional evidence of their relatively advanced
condition. The organization of the western islanders is totemic,
probably in a relatively late stage, there being evidence of a previous
dual organization which has become extinct* The social condition of
the eastern islanders is probably still more advanced, having a terri-
torial basis, with few traces of the conditions of mother-right and
totemism from which they have nevertheless probably emerged. On
studying the kinship system of these two peoples we find different
stages of change in the direction of simplification. In the island of
Mabuiag in the west the distinction between the children of fathers
brother and mothers brother is not present, and the name given to
these relatives is also given to the children of father s sister and
mother s sister. That the absence of the distinction is due to loss,
and not to imperfect development, is rendered probable by the con-
dition of the terms used for the older generation ; here there are
still distinct terms for fathers brother, mother's brother, fathers
sister and mother s sister, but there are definite signs that these dis-
tinctions are becoming blurred, and that the people are on their way
to giving the same name to the relationships of father's sister and
mothers sister, and possibly even to those of father s brother and
mothers brother* In the Murray Islands in the east, on the other
hand, there is still present the distinction between the children of
father s brother and mother's brother ; but here the distinction
between mother s sister and father's sister which seemed to be in
process of disappearance in Mabuiag has completely gone. For the
full evidence on these points I must refer to the articles on ' Kinship *
m the fifth and sixth volumes of the Reports of the Cambridge
Expedition to Torres Straits, I can only say that the evidence is
strongly in favour of the wide connotation of certain kinship terms
in Torres Straits being a product of late change. These changes
would not have to go very much further to produce kinship systems
approaching very closely to that of Hawaii, and thus a strong
supposition is raised in favour of the Polynesian system being also
n product of late change.

If we now turn to Australian systems we find that it is universal,
so fai' as the evidence goes^ to have distinctive names for the four

CLASSIFICATORY SYSTEM OF RELATIONSHIPS 313

kinds of relative of the generation older than the speaker, viz. father
and father s brother, mother s brother, father s sister, and mother and
mother's sister. Similarly, in the next generation it seems to be
almost miiversal, ignoring differences according to age, to have one
designation for father's brother's children and mothers sister's
children and another designation for mothers brother's children and
father s sister's children.

The only exception with which I have met is very instructive
from the point of view which I am considering in this paper. The
exception is found in the case of the Kurnai. In this tribe, which
differs fi*om all other Australian tribes in its mode of social organiza-
tion, there are separate designations for fathers brother, father's
sister, mother's brother and mother's sister, but in the next genera-
tion the corresponding distinctions are absent and the children of
mother's brother and father's sister i-eceive the same names as the
children of father s brother and mother's sister.

In tliis respect the Kurnai system resembles that of the island
of Mabuiag in Torres Straits while it retains the distinction between
father's sister and mother s sister which has disappeared in Murray
Island.

In one place ^ Howitt speaks of the Kumai system as primitive,
though two pages later he expresses doubts about this, Tlie case
seems to be very much like that of the Torres Straits people in
that the social system of the Kumai has a territorial basis with
patrilineal descent, and few anthropologists would doubt that it
represents a late stage in the evolution of Austrahan society. There
can be equally little doubt that the special features of the kinship
system of tlie Kurnai depend on loss of distinctions which once
existed, rather than on a failure to develop distinctions found
everywhere else in AustraUa,

If we accept the view that botli the Kumai and the people of
Torres Straits show us late developments of social organization, we
are confronted with the fact that in these relatively advanced
societies we find variants of the classificatory system w^hich bring
them near to the Hawaian form, though in none of the three has the
generalization reached the degree present in that form.

We now know that the people of Hawaii and other Polynesians
are far mom advanced in social culture than the inhabitants of either
Torres Straits or Australia, and it seems an almost inevitable
' Native Tribes of South-East Amimlia, p. 168.

814

ON THE ORIGIN OF THE

conclusion that the changes which have occurred in the less advanced
peoples have in the more advanced peoples proceeded still further
in the same direction, and have produced the system characterized
by the extremely wide connotation of the relationship terms to which
Morgan gave the name of Malayan.

If we now turn from these regions bordering on the Pacific
Ocean to the islands of the Ocean itself we find evidence pointing,
I think, in the same direction. We find that tlie relationship
systems of Kji and Tonga possess the distinctions between father s
brother and mothers brother and betw^n father's sister and
mother's sister, and they also possess the distinction between the
children of father s brother and mother s sister on the one hand and
mother's brother and father's sister on the other hand. No one can
have any doubt that the people of Fiji and Tonga are in a much
more primitive stage of social evolution than the people of Hawaiis
perhaps the most advanced of Polynesian societies, and though it is
of course possible that the more developed society, so far as general
culture is concerned, may have preserved a more pristine system of
relationships, the association of highly developed general culture and
a late form of relationship system is by fai' the more probable.

So far as I am aware, we have no accounts of the Hawaian
system other than that recorded by Morgan, but an account of the
allied Maori system has recently been recorded by Elsdon Best,^ and
I think that any one who compares this account with those of the
Torres Straits or Fiji can have very httle doubt that we have in the
former a later stage of the Papuan or Melanesian system. It would
seem that just as the Polynesian languages have arisen by simplifi-
cation of those of the Melanesian family, so have the Polynesian
kinsliip systems arisen by simplification of a variety resembling
those found among Papuan and Melanesian peoples at the
present time.

Lastly, let us go to Moi^an's own people, the North American
Indians. Among the systems recorded by Morgan himself we find
some which approach the Malayan system. I will take only one
example. An isolated band of the Iroquois, called the Two
Mountain Iroquois, had a form of the classificatory system in which
the father's brother was distinguished from the mothers brother
(though the two names are singularly alike) ; but the distinction
between father s sister and mother s sister was not present, nor was
' Joum, Anthrop, ImiiL^ 1TO2, vol. xjtxii, p. 186,

CLASSIMCATORY SYSTEM OF RELATIONSHIPS 315

any distinction made between the children of the father's brother,
father s sister, mother's brother and mother's sister. Thus we have
in the case of this Iroquois tribe a system which is rather nearer the
Hawaian system than that of either Mabuiag or Murray Island, each
taken alone. If the definite loss which the Mabuiag system has
undergone were combined with the loss which the Murray Island
system has suffered, we should have before us a system ahnost
identical with that of the Two Mountain Iroquois,

The Two Mountain Iroquois were colonists from the Mohawks
and Oneidas who had settled above Montreal, and if their system is
to be regarded as primitive, we have to suppose that this small band,
who had apparently separated from the main body at no distant date,
had preserved a primitive form, while the main body showed the
usual features of the classificatory system. The system of tlie Two
Mountain Iroquois was collected by Morgan himself, and we may
therefore expect it to be accurate, and it is surprising that Morgan
should have allowed this peculiar system to pass ahnost without
notice, for more attention to it might have led him to re\dse his
opinion that the Malayan form represents an early stage in the
evolution of the classificatory system, and with the disappearance of
the Malayan system as a primitive mode of expressing relationships
would also have disappeared his sole evidence for the existence of
the consanguine family.

The Origin of the Classificatory System in Group- murriage.

In the first part of this paper I have dealt with Morgan's
evidence for the existence of the consanguine family, and I have
shown that so far as the classificatory system of relationships is
concerned we have no evidence for this form of the family. As
I am not here concerned with the general problem of the existence
or non-existence of this form of the family but only with the
evidence for it derived from the classificatory system, I can pass on
to the second part of Morgan's scheme, again premising that I have
only to deal with the existence of the Punaluan family so far as the
evidence for it is derived from the nature of the classificatory
system.

By the Punaluan family Morgan meant a form of the family
characterized by the existence of group-marriage, to use his own
words, * founded upon the intermarriage of several sisters, own and

:n«

fftf IHE OMOCSr OP THB

collateral, with €Mh ocber'a
• • "larriane of MfCrrai

wiv6s, in a gsfmii'. £&
4 rni one dde need not

m A ipftMip,

and ^on the
mn. md mllntonilT with each
6aeh isan be floppned that the
y be of Idzi to one another.

Am Mr. TbMHtti hn dhown^ the eipwaaiiun groupHnarriage has
>wnA very Joeirfy hy rngimlr wtiiM^MMiit will perhaps condoee
tf w# adoft Hr ThwiHi^ MUfcion, thou^ it doea not
with that of Mbrsaa'^ When I usethe expreasian
^gRMqp-marnage \ I shall tfaerefore mean a marriage occurring in a
inity diTidad mt» difciitii gRM^n^ whuliiirr they be clans,
in wIbA dl tbe mat ef one gnmp are the
f ef aB the women of tiie other groapv and all the women
of the first grmqi are the wivea of tim man of the seoond group.
Acr>ording to th» dtfnilion all the hoabanda or wives would be
lalatad m members of the aanna groups and it is in this respect that
Am ddioition may differ from tihat of Mocgan.

The aigmmau ta for the existenee of group-marriage derived from
the elaanifieatofy STStem are bfi^hf as fbUows. Often, but not by
mij means in all forms of the gyatem, a man of one group will apply
tba same term to all the women of another group of a certain
tion which he qipliee to his wife, and conversely aU the
of one group may ap{^y the same term to all the men of
another group and of their own generation which they apply to their
own individual hu^bandSi and it has been argued that these terms
are survivals of a state of society in which there were actual marital
relations between those who used the terms. Secondly, a child of
one group will give the same term to all the men of his father's
group and generation which he applies to his own father^ L e, to aU
thrj«e who under the last heading would in some sj^tems be called
hufiband^ by his mother, and it is supposed that this wide use of the
term * father ' is similarly a survival of a state of society in which all
the men of a certain standing in the opposite group were his potential
fathern. To this argument the objection is made that the child in
all foTniH of the classificatory system gives the same name to the
women of his own group and of the same generation as his mother
as ho gives to his own mother.

lliis objection to the value of the classificatorj^ system as a test
of i>r*^vioim social conditions was recognised by Darwin in liis
refert^nco to the views of Morgan in TAe Descent of Man.^ He

* 1871, vol ii, p. a59.

CLASSIFICATORY SYSTEM OF RELATIONSHIPS 317

remarks * that it seems almost incredible that the relationship of the
child to its mother should ever be completely ignored, especially as
the women in most savage tribes nurse their infants for a long time \
The objection still continues to influence many in their attitude
towards the classificatory system, and the most recent writer on the
subject, Mr. N. W, Thomas, has regarded the objection as a reductio
ad absttrdum of the hypothesis of group-marriage, and has jocularly
commended such a belief in group-motherhood to the notice of
zoologists.

Two quite different answers to the objection are possible. It
may be that there was once a definite term for the individual
relation between mother and child, and that the term became
extended at a later stage of evolution so a^ to fall into line with
other kinship terms. That such an extension of meaning can have
taken place is summarily dismissed by Mr. Thomas as involving
a process for which we have no evidence and for which no reason
can be seen. As a matter of fact, however, as will be apparent from
what I have said in the first part of this paper, people in low states
of culture do extend the meaning of their kinship terms. Relatives
once distinguished may come to receive the same appellation, and
I see no reason to doubt that this process of generalization may have
contributed to extend the connotation of the term ' mother \ The
other answer, however, probably presents more nearly the genesis
of that generalized relationship which we have to translate by that
of mother and child. In such a state of society as that we must assimie
when the system of relationships was in process of development, it is
not probable that the special relationship between mother and child
would have persisted beyond the time of weaning. Let us assume
that the weaning did not take place till the child was three years old ^
and the separation would have occurred before the age at which the
child began to learn the terms of relationship to any great extent
It is even possible that in this early stage of culture the duty of
suckling may have been shared by other women of the group, and that,
at the time of weaning, the cliild might not have been in the position
to differentiate between its own mother and the other child-beaiing
women of the group.

' Op, cit., p. 128.

^ I have asaumed that weaniBg took place at this late age, because this now
happens among many races of low culture, but if it was earlier, my argument Is
only strengthened.

m§

on THE ORIGW OP THE

grow up

•Uata

nnfliijr. We kocfir, boireirwv tl^^ ii^ reblniaif

with paternal dommtt ^ m in the Mnmj Uaiii^m mm i

wHhout luiowiDf UtfttdfrtiierndiD^^ Inflwi

to da wttb §6opAm^ and Ibe eaw la l lwniiiM sot

oeeuiTifiee ot aoch igoonaea in a mlaliiaiy highl]

msisity may help as to uodeniiand tiia ahaence of the knowfedlga of

the poraooality of tha mother at the mueh hmm ategea of aodU

eirolution wiaeh we hare to aaaama at the time of origin of tiia

daarificatory ayntent

Agaitif the auliject of adoptioa, which I have juat mentiooed^
may throw wme light on the matter. The people of the Morray
lalaiidii carry the diatom of adoption to what aeema to us an absurd
extreme^ and ckUdreD are transferred from fionily to family in a
way for which the people can give no adequate reason, nor can any
ader|uate leaaon be found in the other features of the social or
roligiouK inntitutiona of the people* I do not wish to go so far as to
auggefft that thiH custom of adoption may be a survival of a state of
aoeiety in which children were Largely common to the women of the
group so far as nurture was concerned ; but this is possible, and in
any caae this wholesale adoption may help the civilized person to
lUHlerMtiirid tliat people of low culture may have different ideas in
connexion witJi parentage from those prevalent among ourselves, and
that the idea of group-motherliood is not as absurd as Mr. Thomas
Mupposes,

Only one other relationship term raises any serious difficulty,
viz, tho application of the same terms to all the children of the
grovip which aro nppllL^d to own brothers and sisters, but if my line
of argument is accepted to explain * group-motherhood \ the existence
of gron|>-l)rotlun'hood and sisterhood will present no difficulty.

The point whicli I have considered is the most definitely formu-
IiiIimI ohjtH'tion wliicli has been brought against the value of the
f^buHMifirntory HysLein hh evidence in favour of group-marriage. The
oMer objectionH ' weir l>;istrl on the idea that the system is only a
table of UyriuH of adilreHs, a view which by no means removes the
tleceBMity lor a theory of its origin. The tendency of more recent

^ Mclioitniiii, Studks in Aticimt History^ 1876, p, 366.- See also Westermarck,

CLASSIFICATOKY SYSTEM OF KELATIONSHIFS 319

objectors has been to show that the terms of the system are
expressive of status and duties and not of consanguinity or affinity J
I shall return to this point later and will only say now that the view
that the classificatory system had its origin in group-marriage iniphes
that it was in its origin expressive of status rather than of con-
sanguinity and affinity.

Merely to reply to objections raised by others is, however,
hardly satisfying. In the earlier part of my paper I have shown
that we have reason to modify Morgan's scheme in a very funda-
mental respect, and it is now evidently necessary to restate the
mode of the hypothetical origin of the classificatory system in a
condition of group-marriage. Such a statement must be so highly
problematical and must involve so many doubtful features that I am
very loath to undertake the task, I only do so because I believe
it may assist cleaniess in the discussion of the problem if some
definitely outlined scheme has been formulated which may make
clear the points on whicli further evidence is required. My aim will
be to suggest a state of society which would be capable of explaining
the origin of the classificatory system of relationships and at the
same time is not in obvious conflict with what we know of man in
low states of culture.

I shall have to begin by making certain assumptions. First, I
assume that at the time the classificatory system had its origin, the
custom of exogamy was already in existence, and further I aasume,
for the sake of simplicity, though it is not essential to my
argument, that the community possesses only two exogamous
sections, which I will call moieties. We now have so much evidence
of such a dual division of early society that there are few who will
object to this assumption, though my argument would apply equally
well if there were more than two exogamous divisions of the
community.

Further, I assume, again for convenience* sake, that the child
belongs to the division or moiety of its mother. This mode of
counting descent is again so widespread in communities of low culture
that few will quaiTol with this assumption. In the hypothetical com-
munity I assume we have therefore two cnoieties united in group-
marriage, all the active men of one group being the husbands of all
the child-bearing women of the other group. In each moiety four
groups of people would be roughly distinguished ; the active men,
^ Lang, Social OH^im, p, 102 ; N. W. Thomas, op. cit,, p. VZS,

320

ON THE ORIGIN OF THE

the child-bearing women, the elders and the children. The dis-
tinctions between these groups will be fairly clear except in one
case. All that we know of savage ^ciety would lead us to expect
that there would be a sharp distinction between the group of
children and their seniors. The widespread ceremonies of initiation
point to a time when there was a complete change of status at this
period of life, and I assume that the change takes place at a definite
time, i.e. that a boy does not become a man gradually as with us, but
suddenly at the period of initiation. The distinction between child-
bearing and older women would also present no difficulties, and the
chief trouble in imagining the state of society I suggest arises in
connexion with the distinction between the active men and the
elders* If I may be allowed to pass over this difficulty for the
present, we should find in such a society that a child would recog-
nize in his conmiunity people who stand to him in eight different
relations* In his own moiety there would be the group of child-
bearing women to whom he would give a name which was the
origin of that we now translate ^ mother '. Secondly, there would
be the active men of liis own moiety to whom he would give
a name which later came to denote a relationship which we
translate 'mother's brother'. Thirdly, there would be the group
of children to whom names would be given wliich later came to
mean * brother ' and * sister *. Lastly, there would be the group of
elders whose names would have been the origin of the terms trans-
lated * grandfather ' and * grandmother \ In the other moiety there
would be four coiTesponding groups ; men to whom the chUd would
give the name which we now translate * father ' ; the group whom he
would call by the name which came to mean * father's sister ' ; the
children of the moiety to whom he would give a name which later
came to denote the children of the mother's brother and father s
sister ; and lastly there would be the group of elders who would
probably receive the same names as the elders of his own moiety.

Such a state of society would give us the chief terms which we
find in the classificatory system, and new terms would be developed
as the social organization became more complex.

In such a state of society I suppose that the status of a child
would change when he becomes an adult, and that with this change
of status there would be associated a change in the relationship in
which he would stand to the members of the different groups. The
great difficulty in the acceptance of my scheme is to see how the

CLASSIFICATORY SYSTEM OF RELATIONSHIPS 321

relationships set up by these age -groups developed into those
regulated by generations such as we find among most people of
low culture at the present time,

I cannot here attempt to follow out such a development
in any detail, but I think it is possible to see the general lines
on which one almost universal feature of the classificatory system
may have evolved, viz. the distinction between elder and younger,
especially frequent in the case of brothers and sisters. A man
would probably tend to distinguish with some definite ness those
who became adults earlier than himself from those who came
later to this rank ; he would tend to distinguish sharply between
those who helped in his initiatory ceremonies and those to whom he
was himself one of the initiatorSj and tliis distinction between seniors
and juniors would probably bo carried over into the system of
relationships which gradually developed as the group-relations
developed into more individual relations between men and women,
and as the society became organized into generations in the place
of status* or age-groups.

There still exist in various paints of the world societies possessing
age-grades,* which may well be survivals of some such condition of
social organization as that I suppose to have been the origin of the
classificatory system. We have at present no evidence to show what
relation there may be between these age-grades and the systems of
relationships, but it is to be hoped that future investigation into the
system of relationships of some community possessing age-grades
may furnish material for the elucidation of the process by which the
evolution from age-groups to generations has taken place.

What I suppose to have happened is that there were at first
purely group-relationships which received names ; that from these
named relationships the people were led to formulate certain
further distinctions which reacted on the group-relationships and
assisted in their conversion into relationships such as we find to
characterize the classificatory system at the present time.

If I am right in the main lines of the sketch I have just given,
the classificatory system was in its origin expressive entirely of

* For a full account of these age-grades^ see S<jhurtz, AUersklassen und Mantter-
bimde, Berlin, 1902. Unfortunately, Schurtz complicates the problem connected
with age-grades by including in this type of social organization the Australian
matrimonial classes which have probably had an origin very diflferent from that of
true age -grades elsewhere.

322

ON THE ORIGIN OF THE

status, Tlie terms would stand for certain relations within the
group to which only the vaguest ideas of consanguinity need have
been attached. Several recent writers have urged that the classi-
ficatory system as we find it to-day is expressive of status only, and
they have regarded this as a conclusive objection to Morgan's views*
In the attacks made on Morgan's scheme during his lifetime the
objections raised were of a different kind, being directed to show
that the system was merely a collection of terms of address and had
nothing to do with status and duties so far as status implied any
function in the social economy. If Morgan \vere now alive I believe
he would agree to a very great extent with those who regard the
systems as expressions of status and duties so far as their origin is
concerned, though his unfortunate error about the nature of the
Malayan system prevented him from seeing to how great an extent
the terms arose out of purely status relationships. It may be
objected that he called the classificatory system one of consanguinity
and affinity, but he called it this because, whatever may have been
its origin^ there is not the slightest doubt that at the present time
the system is an expression of consanguinity and affinity to those
who use it. I have now investigated the classificatory system in
three commimities/ and in all three it is perfectly clear that distinct
ideas of consanguinity and affinity ^ are associated with the terms.
The correct use of the terms was over and over again justified by
reference to actual blood or marriage ties traceable in the genea-
logical records preserved by the people, though in other cases in
which the terms were used they denoted merely membership of the
same social group and could not be justified by distinct ties of blood
or marriage relationship. There is in these three peoples definite
evidence of the double nature of the classificatory system as an
expression of status and of consanguinity, and there are definite
indications of a mode of evolution of the systems by which they are
coming to express status less and ties of consanguinity and affinity
more.

The evidence relating to the classificatory system brought forward
by most of the recent critics of Morgan has been derived chiefly from
the Australians, and, so far as our existing evidence goes, it would
seem that the status aspect of their systems is more prominent than

ahip.

Mabuiag and Murray laknds in Torres Straita, and the Todas in India,
' By consanguinity I mean blood relationship; by affinity, mamage relation-

CLASSinCATORY SYSTEM OF RELATIONSHIPS 323

in other parts of the world, as would be expected from the very
special development of matrimonial classes among them ; but even
in Australia it is probable that the aspect of tlie systems as
expressions of coiisangiunity and affinity is far more important than
the published accounts lead one to believe. The true relation
between the classificatory system and the actual ties of blood and
marriage relationship can only be properly brought out by a full
application of the genealogical method, and this method has not yet
been applied in Australia.

That there is sometimes a definite connexion between marriage
regulations and the classificatory terms of relationship there can be
no doubt. Thus I have shown elsewhere ^ that the terms used by
Dravidian peoples provide definite indications of the maiTiage of
cousins, which is a feature of their society ; and similarly there is
an evident relation between the classificatory terms and forms of
marriage among the North American Indians.^ When we find
special features of the classificatory system to have had their origin
in speciid forms of maiTiage, it becomes the more probable that its
general features are the survivals of some general form of marriage.

My object in this paper has been to support the view that the
featiu^s of the classificatory system of relationship as we find tliem
at the present time have arisen out of a state of group-marriage,
while pointing out that this system lends no support to the view that
the state of group-marriage was preceded by one of wholly un-
regulated promiscuity. I should like again to msist that it has not
been my object to consider here the problems involved in the growth
of the human family in general, but only to deal with the evidence
provided by the classificatory system of relationsliips.

The classificatory system in one form or another is spread so
widely over the world as to make it probable that it has had its
origin in some universal, or almost universal, stage of social develop-
ment, and I have attempted to indicate that the kind of society which
most readily accounts for its chief features is one characterized by a
form of marriage in which definite groups of men are the husbands
of definite groups of women.

^ Joum. noij, Asiat Soc, 1907, p. 611.
2 See Kohler, op. cit, p. 82.

Y 2

ON PREHISTORIC OBJECTS IN BRITISH

NEW GUINEA

By C. G. SELIGMANN, M.R, and T, A. JOYCE, M.A.

Within the last few years discoveries have been made in
British New Guinea of pottery fragnients and implements of
obsidian and stone, which differ entirely in type from the pottery
and implements used at the present time by the inhabitants of the
localities in which the finds were made.

The majority of these objects have been brought to hght in the
course of prospecting or mining operations ; some have been foimd
a considerable number of feet below the surface ; others have been
picked up on the surface of the ground or brought in by natives,
who could, however, give no information concerning them. Thus
tlie objects discovered are all truly prehistoric in the sense that in
e;ich find objects occur concerning the origin and use of which
nothing is known by the inhabitants of British New Guinea at the
present time ; there is, however, in no case reason to attribute any
great age to the specimens found.

The discoveries mentioned fall naturally into four classes, viz, :
(i) Obsidian implements,
(ii) Stone implements,
(iii) Engraved shells,
(iv) Pottery.

The majority of the discoveries have been made on or near
the uorthern coast, between the Mamba river and Colhngwood Bay,
but single obsidian objects have been obtained fi'om Misinia in the
Louisiade archipelago, from Goodenough Island and from Murua;
while a peculiar type of prehistoric pottery is found on the small
island of Dauko, off the south coast of the Possession and about

326

ON PREHISTORIC OBJECTS

four miles from Port Moresby, The sites of discoveries are marked
with a + on the accompanying map (fig. 1).

Fig. t

Obsidian Implmiients.

Before describing the unusual obsidian implements which form
the first class of prehistoric objects, it is necessary to consider briefly
the occurrence of obsidian in the villages of south-eastern British
New Guinea at the present day* Small fragments are found mixed
with the sliingle on which are built the coastal villages of Bartle
Bay^ an indentation in the large hollow of the coast which faces the
D'Entrecasteaux group and constitutes Goodenough Bay. These
fragments were until recently used for scarification for medical
purposes, and the blocks from which they were struck were stated
to have been brought from Goodenough Island for this purpose ;
but it was said that no laj^er fragments were in existence, that
implements were never made of obsidian, and that no one had ever
heard or thought of applying it to any use of this kind. At Waga-
waga, in Milne Bay, tiagments of obsidian, formerly used for bleed-
ing and scarification, though less abundant, were not uncommon^
and here they wore said to have been obtained from a place called

IN BRITISH NEW GUINEA

327

Hiliwau, described as near East Cape, where, according to a some-
what doubtful statement, obsidian boulders were found in the
jungle. One of us (C< G. S,) was shown a lump of obsidian about
as big as an orange from which small masses had evidently been
struck, and also a rather large piece of a volcanic glass, of a
brownish red colour, said to have been brought from the same
locality. But, again, it was denied that implements of obsidian had
ever been made either at Wagawaga or elsewhere, and the same was
said at Tubetube in the Engineer Group, where the fragments of
obsidian used for medical scarification were formerly imported from
Duau, the largest island of the DEntrecasteaux group.

The localities mentioned were those in which a considerable
amount of work was done by the members of the Daniels Ethno-
graphical Expedition, but a flake of a brownish volcanic lava,
suggesting a pitchstone rather than obsidian, was found on Gawa,
one of the coral islands of the Marshal Bennet group, and the piece
of worked obsidian shown in pL ix, fig. 6 was picked up on
Murua. Practically, then, fragments of obsidian have been found
wherever search has been made in the south-eastern portion of
British New Guinea and its archipelagoes, but nowhere, as far as
our present knowledge extends, is there any legend or trace of a
belief that it was ever worked to form such implements as are
shown in pL viii, figs. 1, 2, 6, and 7.*

The most striking of the prehistoric obsidian objects is the
beautifully finished axe or adze shown in pi viii, fig. 2, and now in
the possession of Mr. David Ballantine, who added the modern haft-
ing,* The dimensions are as follows :— greatest length from cutting
edge to end of tang, 183 mm., greatest breadth, 215 mm. It was
found in a creek in the Yodda valley, below the surface of the ground,
together with the stone pestle and moi-tar described below (p. 329).

* Dr. Rudolph P6ch, who has apent some timo stmljing the Komfi of Colling-
wood Bay, informs us that he found many fragments of obsidian at Kainu, where
he dug into one of the mounds which had not previously been distur1>ed.
Dr. Fdch also states that he found no implements of obsidian in use among the
folk of Colling^vood Bay and the Cape Nelson Peninsula, although obsidian cores,
from which smAlI fragments were struck when required for shaving or acarifica-
tion» are in common use.

- We take this opportunity of thanking Mr. Ballantine for permission to
publish the photographs of implements in his collection shown in pi. viii, figs.
1 and 2 ; further, the flake shown in pi. viii, figs. 6 and 7, placed in the British
Museum by one of us, was obtained through him*

328

ON PKEHISTORIC OBJECTS

Tills adze-blade is made ftom a single lai^e flake, and is roughly
lozenge-shaped witli a rounded cutting edge ; on the shaft side, as
hafted, it has been worked to forai a short tang.

The flake, of which two views are shown in ph viii^ figs. 6 and 7,
was obtained on Goodenough Island, where it was brought for trade.
It had been insecurely, and quite recently, lashed to the ends of
two of the long thin spears typiciil of the D'Entrecasteaux group.
On one face there is a median ridge, in part double ; on the other
face there is a distinct bulb of percussion at the end remote from
the point Length 113 mm. (BritiBh Museum, No. 1906. 10-14. 9).

The spear head illustrated in ph viii, fig. 1, length 135 mm.,
was found by a miner, when sinking a shaft on Misima, at a depth
of 4 metres below the surtaee. It is formed of a single flake
triangular in outline, one end worked to form a broad short tang.
One face has a double ridge, the other is plain. The small stone
adze-blade mentioned below (p. 329) was found in the same shaft
some 5 metres lower.

Tlie last specunen of obsidian is shown in ph ix, fig. 6 ; it is
a small flake picked up by one of us (C. G. S.) on Murua at or
below high-water mark in Wanai Bay. On the same site was found
a considerable number of well-weathered chips and flakes of typical
Suloga adze-stone.^ This flake is roughly quadrangular, and lias a
dull surface. One face is plain with a well-marked bulb of per-
cussion, the other is flaked. L, 26 mm. (B. M. 1906. 10-13. 33).

Sionc Impkments,

The two most interesting of the stone uiiplements are the pestle
and mortar sho^vn in pL ix, fig. 7, which were found in the Yodda
valley, in the same creek as the obsidian axe already mentioned.

Mr. C. A. W. MoncktoUj the Eesident Magistrate of the north-
east district, mentions the discovery of these in the following terms :
*A remarkable pestle and mortar , . . have been fomid by some
miners in gold workings at a depth of 12 feet below the surface
in the Yodda valley. The mortar, which with the pestle weighed
66 lb<, was roughly ornamented with barbaiic carving . . . the

* Suloga is the nam© of the site whence was obtained the stone which, in
the form of ad7,e-hlades, passes in trade for many miles on both the northern
and southern coasts of the Posseasion* In fact^ on the southern coast Suloga
blades have reached at least as far as the Papuan Gulf*

IN BRITISH NEW GUINEA

329

pestle aiid mortar were discovered in the same creek as an obsidian
battle-axe given by me some years ago to the Hon. David
Ballantine, and both would appear to be relics of a forgotten race*
No native to whom the recently found articles were shown could
make any suggestion as to their ori^nal use or purpose, and all
agreed that they were not the work of any now existing tribes/^

The mortar has been cut from pale and rather soft stone ; it is
oval and measures 445 x 386 mm, ; the bowl-shaped depression
measures 295 x 285 mm. and is 60 mm, deep at the centre ; round
the depression runs a low ridge, which again is sun*ounded by
fomieen large irregular knobs fringhig
the edge of the mortan The pestle,
which is cut from hard greyish stone,
is 165 mm. long, and resembles a
slightly elongated pear.

Several similar pestles, one of
them a quite remarkable piece of
carving in stone, have been found in
this neighboin-hood.-

A highly patinated adze-blade, of
which the outline is shown in fig. 2,
was found by a miner in the same

speai-

Fig. 2,
shaft as the obsidian
head described above, some 9 metres below the surface.

The remaining stone prehistoric objects come from Rainu in
Collingwood Bay which, as Mr. Monckton says, is evidently ' an old
village site of a forgotten people '. Here, too, were obtained frag-

' British New Guinea Annual Eepoti, 1903-4, Appendix B, p. 31. Mr.
Monckton figures the mortar and pestle which owing to his wise generosity
are now in the Britiah Museum (B. M* 1904, 11-23. 1), but the reproduction
is unfortunately by no means a good one.

* Since this was written Mr. C. W. de Vis has published the description of
a stone pestle found on Murua under three feet of gravel at the bottom of an
extinct river bed from which were obtained * the fossil bones of dugong, turtle,
and crocodile'. The pestle is made * of diabase or diorite^ the rock which ... is the
prevailing geological feature of the island *. In shape it is described as resembling
* a short hyacinth glass, with a bulb of the plant in its UBual position. Its base is,
as it should be, gently and regularly convex ; its conical body suddenly dilated
above into a thick collar ... an obtusely conical knob surmounting the collar,
testiiies to an impulse of the artistic faculty. • . . Its dimensions are these: —
Total height, 109 mm. ; diameter at base, 88 mm. ; at neck, 43 mm. ; of collar,
64 mm." (Annals of the Queensland Museum^ No. Ivii, 1907, p. 12.)

IN BRITISH NEW GUINEA

881

smaller lighter immigrant Papuo^Melanesians^ who have pushed up
the coast from the east.

A stone fragment (B. M. 1905, 2-9, 342) of peculiar form was
found here ; the material appears to be a coarse-grained not veiy
hard sandstone ; from the fact that it is cur\ ed it appears to be the
centre of the base of a vessel. On the exterior is a short cyhndrical
projection, possibly a foot. The interior is smooth and appears to
have been much worn by friction. L. 160 mm. It seems not
unreasonable to conclude that this is a part of a mortar similar to
that already described.

The adze-blades found on the Kahiu site are of the stone
ophicalcite,^ which is quarried (probably by the Doriri) somewhere
on the northern aspect of the Goropu range, and traded down the
Wakioki River to the Maisin, and dowm the Musa river to the more
northerly tribes of Cape Nelson, But the four blades that we have
examined, including those now in the British Museum, are all
smaller and less heavy stones thau those in present use, of which we
have handled a considerable number.^ PL ix, figs. 1 and 2, shows
two of the ophicalcite adze-blades from the Rainu excavations ; one
of them is thin and flat ; its outline is an irregulai^ oblong ; one end
is ground on both sides to form an edge. Length 120 mm. The
other is thicker, roughly triangular in outline ; both sides are
pohshed and bevelled to form a cutting edge at the base of the
triangle. Length 50 mm.

A fragment of a discoid clubhead (B< M. 1905. 2-9. 327), cut
from what is probably the same stone, though darker and harder,
was also found in the excavations. It shows a portion of the usual
biconical central perforation.

' We are indebted to Dr. J. E. Marr for kindly identifying the stone from
which thei^e adzes are made.

* We take this opportunity of gratefully acknowledging the assistance we
have received from Mr. G. O* Manniug, the Kesideut of the North Eastern
Division in which CoUingwood Bay is situated. Not only has he answered
many questions and traced the distrihution of ophicalcite adze-blfides on the
northern coast, hut he has sent to this country a carefully lal>eUed series of
adzes collected Ijetween CoUingwood Bay and the northern boundaiy of his
division in the neighbourhood of Cape Endidadere, The Ijest of these adze
blades are now in the British Museum , where they constitute iiuml>ers KiOO.
10-14. 3, 4, 5, 6, 7^ and 8. The numbers of the prehistoric ophicalcite adze-
blades from Kainu in the national collection are 1906. 10-14. 2 and 3 and 1005.
2-9, a29.

332

ON PBEHISTORIC OBJECTS

Fig. 4,

Engmveil Shells.

Five specimens of these are known to us. One is the armlet
cut from a Conus shell ah-eady mentioned; another is an entire
eonus shell ; the remaining three have the flat spire removed, and
one of these has the interior removed also.

The armlet shown in %• 4 is annular, formed of a cross-section
near the base of the cone. The exterior, now much weathered, has
been ornamented with engraved concentric arcs and
diagonals. Diameter 60 mm., depth of ring 10 mm,

PL viii, fig. 6, represents the complete conus shell,
brownish in colour and ornamented with irregular
lozenge patterns arranged in panels (B. M. 1905*
2-9. 336).

Figiu'es 3 and 4 of pL viii are shells with the spire removed,
ornamented with bands of incised continuous scroll patterns. In
the specimen shown in fig- 4 (B. M* 1905. 2-9, 337) the interspaces
are filled with a sort of hatched leaf pattern ; in that illustrated in
fig. 3 (B. M. 1905. 2-9, 338) the bands of ornament appear to centre
round a conventional human face which is extremely suggestive
of the art of the Elema tribes of the Papuan Gulf,

The remaining specimen (not figured) of which the spire,
columella, and, indeed, the greater part of its ventral surface have
been removed, is ornamented with continuous scroUs similar to
those on the two last specimens described.

These engraved shells are perhaps the most puzzling of the pre-
liistoric objects that have so far been found in British New Guinea.
Not only is engraving on shells not practised elsewhere in the
Possession, but that portion of the cone sliell which remains when
its spire has been removed to be rubbed down to form a pendant, or
when an armshell is made from the broad end of the cone, is never
used as an ornament.^ But of these prehistoric engraved shells two
of the four known specimens show that this remaining part of the
shell was carefully decorated in a way which suggests that these
shells must either have been highly valued ornaments, or else have
constituted a class of object which has now ceased to exist in the
Possession. Indeed, the question admits of being considered on a

* We may, however, refer to the designs, usually totemiHtic, soratdied on
fihells of the large pearl oyster {Mtkagrirui margaritifera) in Torres Straita
Cf, Camh. Univ, Expedition to Torres Straits^ vuL v, fig. 21, p. 169,

m BRITISH NEW GUINEA

wider basis tlian is implied by the statement that nothing like these
shells has previously been found in British New Guinea, and we
' believe it may be stated that engraved cone shells or conns armlets
have not hitherto been reported from Melanesia. Although three
of the specimens under consideration are so much weathered as to
cmmble easily, cone ehells when fresh are extremely liard. The
question then arises how the engraving was done. The ophicalcite
blades found on the Rainu site with the cone shells ^vill not scratch
a good specimen of the modern cone shell armlet. Stone * quarried *
at Suloga will scratch such an amilet, though not readily; we,
however, found that fragments of obsidian picked up in the villages
of Goodenough Bay readily scratched an armshell, leaving a clean,
weUniefined furrow such as the Rainu shells must have exliibited
when first engraved. The condition of these shells does not, to
our mind, oflFer any solution to the probleni of their age ; their
surface is so weathered as to be scratched easily with the thumb-nail,
but we cannot say whether this indicates an age greater than a
couple of decades, considering the fact that the shells have been
buried at no great depth in damp soil in a hot eUmate with a pro-
longed rainy season*

Pottery.

In 1905 Mr. Monckton gave to the British Museum a large
collection of pottery fragments from the Rainu site. The chief and
obvious interest of his gift Ues in the fact that this collection
consists of fragments of vessels far superior, in construction,
symmetry, and ornamentation, to anytliing which is made in the
neighbourhood of Rainu at the present day. From the fragments
it is clear that the vessels were mainly of large size, and the sides
average about 10 mm. in thickness ; the pottery is liard, often ex-
tremely hard, and usually shows a reddish or black exterior. Most
of the pots, to judge from the fi-agments, were either large bowls,
hemispherical or more than hemispherical, or shallow circular dishes,
some encircled with a broad flat hp, some with low sides rising
more or less abruptly from the curved bottom.*

^ This prehistoric pottery is moat nearly approached by certain large hemi-
spherical vesaels in which the cleaned hones of the Miima dead are exposed
in rock shelters ; some of these bowls collected by the Daniels Ethnographical
Expedition and now in the British Museum (Nos* 1906. 1(K-13. 34 to 38) are
420 mm. wide and 230 mm. deep. Their rims are often ornamented or im-
pressed with inci&ed patterns and the latter are generally present forming a circle

334

ON PREHISTORIC OBJECTS

Most of the fragments figured belong apparently to vessels of
the first description ; on pL xi, fig. 10, and pi. xii, fig. 9, are examplea
of the two varieties of the second.

The majority of the pots have a lip surrounding the rim ; this
lip may be rounded, and curve outwards, as in pL x, fig. 5, &e.,
or it may be flat and project abruptly, as in pi. xi, fig. 5, &c.
Where there is no rim, there is frequently a handle, which may be
of the * ribbon ' type, pi. xii, fig, 1> (fee, or of the type sho^vn in
pi xii, fig. 4, familiar from the so-called *food vessels' from
Northern British barrows of the bronze age. The resemblance is
strengthened by the tendency shown in both cases for these handles
to become mere ornamental excrescences and to appear in more
than one row. In some cases the vessel is encLrcled by a flange^
moulded, as pi. xii, fig, 14, or with pierced work, pi, xi, fig. 3, &c.

The ornament is extremely varied, and is chiefly incised or
impressed ; the most frequent designs are punch-marks, spirals,
concentric arcs, and circles, string pattern, groups of straight lines
arranged in triangles, meander patterns, and so forth. Occasionally
a zigzag is obtained in relief by means of a series of triangular
impressions (pi. xi, fig. 9), Perhaps the most striking form of orna-
ment is that obtained by means of open work. This decoration in its
simplest fomi appears in pL xi, fig. 4, where round holes are punched
in a moulded flange. These holes may well have served some useful
purpose, e. g. for suspending the vessel. In fig. 3 of the same plate
the apertures become more essentially ornamental ; in figs. 2, 6, 6 the
openwork has invaded the lip, and in fig. 1 of this plate and pi. xii,
fig. 16, ornamental apertures occur in the side of the pot itself.

Next to the open work, perhaps, in interest is the remarkably
perfect spiral shown in pi. x, fig. 3.

Another form of decoration is represent' J in the apphed circles,
meanders, and bands, which often accompany the incised ornament.
These applied bands, &c., are sometimes plain and sometimes in-
dented. In connexion with this form, attention may be called to the
fine appUed spiral in pi. x, fig, 2. Ornamentation is usually confined
to the exterior, exceptions such as pi. xi, figs. 8 and 16, being rare.

Among the pottery fragments were found three which must
evidently have been the necks of bottle-shaped vases. Two of
these are plain as fig, 13 of pL x, the other (fig. 12) has transverse

immediately below the rim. Captain Barton informs us thai these howls are made
at Kwatota, an island of the Amphlett Group.

IN BRITISH NEW GUINEA

335

ornamental ridges in low relief. A glance at the interior surface
of fig. 13 shows that it was made by coiling. It is noteworthy that,
as far as the country is known, no vessels with necks are made or
used in British New Guinea. The nature of the pottery object
shown in fig, 11 of plate x is uncertain. It somewhat resembles
the fragment of a spindle whorl wliich has been broken longitu-
dinally ; possibly it is the weight of a pump drill. In pL ix,
figs. 4 and 6, are shown two pottery club heads which also come
from the Bainu site. These are presumably ceremonial, as is the
modern wooden club cut from the solid, of which the head is shown
.for purposes of comparison in fig, 3 of the same plate. The latter,
although obtained at Port Moresby, was said to come from the
mountains far inland.

The last series of prehistoric pottery (pL xiii and fig. 5) con-
sists of a number of fragments collected by one of us (C. G. S.)
on the island of Dauko, some four miles fiom Port Moresby. Dauko
is a low coral island covered with rank grass and bearing a few
scattered trees and bushes of the common coast-loving plants of the
district. It is at present uninhabited, and all that could be dis-
covered concerning any previous population is a legend that, for
a brief period, perhaps about four generations ago, the folk of the
Port Moresby villages, together usually termed Hanuabada, migrated
to Dauko during a severe epidemic and lived there till the disease
had spent its force. But the natives assert that this pottary cannot be
related to this short immigration, and, indeed, the style of ornament
upon these fragments differs entirely from that found among any of
the Motuoid tribes, the pot-makers of the whole Central Division.

The vessels of which fragments were collected differ in type
and decoration from those at Rainu, No trace of a handle is found
on any of the fragments, and in only one case is the pot furnished
with a lip {pi, xiii, fig. 3). This lip is flat and broad and expands
abruptly outwards from the side of the pot, the ornamentation being
on the upper surface of the lip. In all other cases the fragments
show that the pots were of the same type, circular, with a perpen-
dicular or slightly incurved rim about 32-40 mm. deep, below which
the sides take an abrupt bend inwards. The ornamentation, which
in all cases but one seems confined to this rim, is of one kind only,
incised. Bands of lines, simple hatching, cross-hatohing, zigzags,
meanders of string pattern and arcs, are all represented, and the
decoration is more minute and less bold than that of the Kainu

336

ON PREHISTORIC OBJECTS

pots and, in consequence, the general effect of the vessel must have
been far less striking, Tlie exception, fig. 6 (text), mentioned above
is the fragment of the side of a vessel which has been ornamented
with broad bands of a reddish pigment. This, with another frag-
ment from the same site (pL xiii, fig. 3), is the only example collected
from any locality which shows an attempt to apply
coloured decollation to pottery.

The almost perfect symmetry exhibited by the
prehistoric pottery as a whole — ^to judge from the larger
fragments — might give rise to the question whether the
use of the wheel were known. But apart from the
fact that the wheel is unknown in Melanesia, the women
Fig, 5. Qf the Motu stock at the present day make narrow-
mouthed vessels, in some instances with bodies approximately
spherical, of perfect symmetry by the simple expedient of giving
an occasional turn with one hand to the board or fragment of old
pot upon which the lump of wet clay is supported.

DESCKIPTION OF PLATES VIII-XIU (P(yprERY)

PLATE VIII

Fig. 1. Obsidian spearhead j found at lilisima (see p. 328).
FiQ. 2. Obsidian axe-blade ; found in a creek in the Yodda valley (see p. 327)
Figs. 3» 4, and 5. Engraved Camis shells ; found at Eainu (see p. 332).
Fioa. 6 and 7. Obsidian iake ; obtained in Goodenough Island (see p. 328).

PLATE IX

Figs. 1 and 2. Ophicalcite adze-bhides found at Rainn (see p. 331}»
Flo. 3, Wooden ceremonial club ; obtained at Port Moresby (see p, 335),
Fioa. 4 and 6, Pottery club-heads ; found at Rainu (see p. 335),
Fig* 5. Obsidian flake ; found at Wanai Bay, Munia (see p. 328).
Fig. 7. 8tone pestle and mortar ; found 12 ft, below the surface of the ground
in the Yo<Ma valley (see p. 328),

PLATE X

Fig, 1, Pottery fragment ; part of rim of vessel, reddish surface, grey body ;
the side of the bowl expands out^vard to form a broad rim of which the horizontal
surface is decorated w4th incised spirals^ and^ at the edge, with a shallow groove
between two rows of diagonal punch marks. The external edge of the rim appears
to have been moulded. 1905, 2-9. 193.*

^ This and the subsequent numbers refer to the British Museum Registra-
tion Catalogue.

IN BRITISH NEW GUINEA

337

Fxo. 2, Pottery fragment ; greyish-buify ornamented with an applique spiml
in reUef. 1905. 2-9. 302.

Fia. S. Pottery hugment ; part of rim of vessel, reddish grey, ornamented
with an impressed spiral and, at the edge of the rim, a row of shallow circular
impreaaions. The surface of the interior is ribbed horizontally* 1905. 2-9. 194.

Fio, 4- Pottery fragment ; part of rim of vesael, reddish surface, grey body ;
aide ornamented with incised concentric circles ; the rim expands abruptly outward
to form a lip, curling slightly downwards. 1905. 2-9. 203.

Fia. 5, Pottery fragment ; part of side of vessel, greyish-buff, with fluted rim
ornamented here and there with small knobs. Ornaniented with series of im-
pressed concentric circles outlined with punched dots, and with meanders in relief.
A shallow groove encircles the pot immediately below the rim, and below this is
a line of punched dota 1905. 2-9. 214 a,

Fig, 6. Pottery fragment ; reddish grey, with two small applique circles in
high relief outlined with punched dots. 1906. 2-9. 308.

Fio. 7- Pottery fragment; brownish black with curved line in high relief,
below which is an impressed concentric chevron pattern. 1905. 2-9. 298*

Fig. 8. Pottery fragment ; greyish red, ornamented with curved indented
band in high relief. 1905. 2-9. 297.

Fia. 9. Pottery fragment ; part of rim of vessel with everted lip^ below which
is an indented band in relief. Below this again is a band of diagonal lines of
incised string pattern. The edge of the lip is similarly ornamentodt but the
nmrking is very faint, 1905. 2-9. 268.

Fio. 10- Pottery fragment ; part of rim of vessel ; about 3 cm. from the edge,
the vessel is encircled by an indented ridge ; above this, by a meander of impressed
string pattern. 1905. 2-9. 258.

Fio. 11. Pottery fragment; apparently part of an object in shape a double
cone, pierced vertically by a large aperture. The more depressed conical portion
is ornamented with bands of impressed lines outlined with punched dots ; the
more elevated conical portion with impressed lines encircHng it spirally.
1905. 2-9. 315.

Fig. 12, Pottery fragment ; spout of a vessel, red pottery, cylindrical, with
a low flange about 5 mm. from the edge and one or more raised bands, where the
neck meets the body.

Fio. 13. Pottery fragment ; spout of a vessel, reddish brown, quite plain,
cylindrical.

PLATE XI

Flo. 1. Pottery fragment ; part of rim of vessel^ reddish -grey pottery,
expanding outwards to a rounded lip immediately below which is a row of 11 holes
between two hands, with two vertical lines in low relief running up over the
edge of the lip. Below the holes are three ridges in low relief, the second under*
lined with punch marks ; the rest is decorated with impressed herring-bone pattern.
1905* 2-9. 248.

Fio, 2. Pottery fragment ; part of rim of Teasel, pale-buff pottery, lip
expanding outward and pierced with four holes (three only complete), running
diagonally from the upper surface of the lip to the under-side on the exterior ;

338

ON PREHISTORIC OBJECTS

the upper surface of the lip, where not ornamented with holes, is grooved
longitudinally* The esteriar is ornamented with lines of punch marks.
1905. 2-9. 246.

FiQ. 8. Pottery fragment ; part of rim of vessel, reddish pottery ; at the edge the
exterior is encircled by a band of diagonal lines of impressed string pattern ; below
this is a broad flange^ at right angles to the body, with large triangular perforations^
between which runs a double zigzag of impressed string pattern. 1905. 2-9, 239,

Fig. 4, Pottery fragment ; part of a rim of a vessel, reddish-grey pottery j
20 mm* from the edge the vessel is encircled with a broad flange with moulded edge,
perforated vertically with circular holes- The edge of this flange is ornamented
with impressed diagonal string patterns. 1905. 2-9. 244.

Fio. 5, Pottery fragment ; part of rim of vessel, reddish-grey pottery, with
abruptly everted Up, pierced with oblong apertures between rows of punched dots.

Fig* 6. Pottery fragment ; part of lip of vessel, with a double row of
triangular apertures apex to apex alternately, and on the rim a similar pattern
punched in miniature. The edge of the lip ornamented with incised diagonals.
1905, 2-9. 241.

Fig. 7. Pottery fragment; part of rim of vessel, reddish pottery, edge
flattened at the top and forming a lip on exterior and interior, upper surface
ornamented with double row of punched dots ; edge of exterior lip with punched
triangles.

Fig. 8. Pottery fragment ; part of rim of vessel, reddish pottery, everted lip
ornamented on the inner surface with horizontal lines of incised dots interrupted
at intervals by pairs of transverse ridges. The exterior of the pot is decorated
with series of impressed concentric circles outlined with incised dots. 1905.
2-9. 216.

Fro. 9* Pottery fragment; part of rim of vessel, reddish-grey pottery,
abruptly everted lip, the upper surface of which is ornamented with three longi-
tudinal lines of pattern, a zigzag in relief obtained by punching down the
background, between two row^s of incised dots. 1905. 2-9, 288.

Fig. 10. Pottery fragment ; part of rim of vessel, greyish-buff pottery,
abruptly everted lip, the upper surface ornamented with deeply incised herring*
bone pattern. 1905. 2-9. 282.

Fig. 11. Pottery fi^agment ; part of rim of vessel, rounded edge, below which
la a ridge in low relief ornamented with diagonal lines of impressed string
pattern ; below this the body ornamented with incised horizontal lines. 1905. 2-9,
148.

Fig. 12, Pottery fragment j part of rim of vessel, reddish-black pottery ;
round the edge on the exterior runs a zigzag in partial relief, obtained by punching
down the background ; below this panels of rough chevron pattern, incised.
1905. 2-9. 250.

Fig. 18. Pottery fragment ; part of rim of vessel, black potterj^ ; round the
edge on the exterior runs a band of incised diagonal lines grouped in triangles.
1905. 2-9. 229.

Fig. 14. Pottery fragment ; part of rim of vessel, reddish pottery, lip slightly
incurved, upper surface, which slopes downwards toward the exterior, ornamented
with incised diagonal lines, with a row of dots along approximately every other
one ; below, a plain ridge encircles the exterior. 1905. 2«9< 228.

IN BRITISH NEW GUINEA

339

Fig. 15. Pottery fragment; part of rim of vessel, reddish pottery, with
sharply everted lip, ornamented along the edge with impressed dots ; body
ornamented with lines of similar dots and impressed horizontal lines. 1905.
2-9. 218.

Fio. 16. Pottery fragment; reddish pottery, part of aide of a vessel, the
interior ornamented with decoration of impressed string pattern arranged in
panels, 1SK)5. 2-». 307,

Fig. 17. Pottery fragment ; part of handle^ reddish pottery, approximately
eircular in section^ exterior ornamented with three longitudinal indented ridges.
1905. 2-9. 311.

Fig. 16. Pottery fragment; part of rim and side of vessel, greyish-black
pottery, ronnded in-curved lip ornamented with longitudinal grooves; below
on exterior an impressed line of cord pattern, below, a broad band of incised
diagonals arranged in triangles, and below again a series of incised horizontal
lines, 1905. 2-9. 187.

PLATE XII

Fig. 1. Pottery fragment ; part of rim of vessel, reddish pottery, rounded lip
projecting outwards ; about 4 cm. lower runs a horizontal ridge ; the lip and ridge
at intervals expand so as to meet and form a loop handle ; the lip and the space
between lip and ridge ornamented with incised dotted meanders* 1905. 2-9. 4.

Fig. 2. Pottery fragment ; part of rim of vessel, roddish-black pottery ; loop
handle similar to fig. L 1905. 2-^9. 2.

Fig. 3. Pottery fragment ; part of rim of vessel, reddiah grey-black pottery,
small loophandle as in fig. 1, upper surface of lip ornamented with longitudinal
groove ; the triangular projection which forms the handle decorated on upper
surface with three vertical holes and a double groove following the outline ; body
om&mented with horizontal incised lines. 1905, 2-9. 82.

Fia, 4. Pottery fragment ; pari of rim of vessel, reddish«black pottery, rounded
lip grooved longitudinally ; below on the exterior is a triangular projection, the
apex meeting the apex of a similar projection below it to form a loop handle ;
the upper surface of the former ornamented with grooved diagonals, the edges of
both with impressed string pattern. Body ornamented with longitudinal grooves.
1905. 2-9. 40,

Fio. 5. Pottery fragment; part of rim of veasel, reddish pottery, rounded
lip« at one point a projection similar to the loop handle of fig. 4, but imperforate.
Below the rounded edge is an indented horizontal band in relief. 1905. 2-9. 75,

Fig. 6. Pottery fragment ; part of rim of vessel, reddish pottery, rounded
edge, 25 mm. from which is a small vertical applied loop handle* Below this the
body is ornamented with impressed chevron pattern. 1905, 2-9. 289.

Fig. 7. Pottery fragment ; part of side of vessel, greyish-huff pottery ; along
a ridge are a number of small applied projectionsj simUar to the loop handle of
fig* 6, but imperforate.

Fig. 8. Pottery fragment ; part of aide of vessel ; reddish potteiy, similar to
last. 1905. 2-9. 292.

Fig. 9. Pottery fragment ; part of rim of vessel, reddish pottery, lip pm-
jecting outwards ; 40 mm. from edge the sides bend abruptly inwards, the angle

Z 2

340

ON PREHISTORIC OBJECTS

being marked by a well-defined ridge ; above this runs a less prominent ridge^ con-
nected with the first at intervals by aeries of three applied knobs, similar to
those in figs. 7 and 8, but smaller. 1905. 2-9. 287,

Fio, 10. Pottei7 fragment ; part of rim of vessel, dark-reddish pottery^ lip
projecting ontwards and grooved longitudinally on the upper surface. These
grooves ai-e interrupted at one point by a series of six small rounded knobs along
the exterior edge of the Lip. Below the lip run two shallow grooves and an incised
line. 1905. 2-9, 286.

(It may be noticed here that the loop handle ^ appearing in its most perfect
form in fig, 1 appears to pass through various stages of degradation^ becoming
small in fig. B, imperforate and meaningless in figs. 7 and 8, and a mere decorative
excrescence in 9 and 10.)

Fig. 11. Pottery fragment; part of rim of vessel with everted lip, reddish
pottery ; below is a horizontal applied band in relief with losenge-shaped indenta*
tions. 1905. 2-9. 272.

Fig, 12, Pottery fragment ; part of side of vessel, reddish pottery, with semi-
circular discoid projection, 1905* 2-9* 293.

Fig, 18. Pottery fragment ; part of rim of vessel, reddish pottery, rounded
lip projecting outwards^ ornamented with horizontal grooves interrupted by
impressed mouldings, 1905. 2-^9. 280.

Fig. 14. Pottery fragment ; part of rim of vessel, reddish pottery, flattened
lip projecting outward, 20 mm. from which is a prominent ridge with impressed
mouldings. 1905, 2^9. 251.

Fig. 15. Pottery fragment ; reddish pottery, a short circular rod expanding
somewhat abruptly at one end, where it has been broken off ; possibly the leg of
a vessel. 1905. 2-9, 316.

Fig. 16. Pottery fragment ; part of rim of vessel ; black pottery, square
edge with single broad groove, the sides apparently pierced with a row of large
circular apertures, and further ornamented with groups of diagonal impressed lines,
and lines of small incised dots. 1905, 2-9. 306.

PLATE Xin

Fig. 1. Pottery fragment ; part of rim of vessel ; SO mm. from edge the side
bends abruptly inwards. Along the edge runs a horizontal hand of incised lines
with a row of scallops along the lower edge. Immediately above bend runs a
similar band without scallops, L. 76 mm.

Fig. 2. Pottery fragment ; part of rim of vessel, reddish pottery ; along the edge
runs a band of incised lines, and 23 mm. below is a similar band ; between is a band
of chevron pattern, the chevrons composed of four or five incised lines. L. 83 nam.

Fig, 3. Pottery fragment ; part of rim of vessel of pale-buff pottery, with
traces of red pigment on the surface. The vessel had an abruptly everted lip
ornamented on the interior with vertical incised lines arranged in panels between
bands of horizontal lines. L. 50 mm.

Fig. 4. Pottery fragment j part of rim of vessel, pale coarse pottery ; 30 mm.
from the edge the side bends abruptly inwards ; above this as ornamentation is an
incised indented line surrounded hy bands of incised straight lines. L, 92 mm.

IN BRITISH NEW GUINEA 841

Fia. 5. Pottery fragment; part of rim of vessel, reddish • grey pottery;
18 mm. from edge the side bends abruptly inwards ; along the edge on the exterior
is an incised band of hatched contiguous triangles, vertices downwards. L. 86 mm.

Fia. 6. Pottery fragment ; part of rim of vessel, coarse pale pottery, ornamented
with parallel bands of incised cross-hatching and incised meanders. L. 55 mm.

Fia. 7. Pottery fragment; part of rim of vessel, of coarse buff pottery;
28 mm. from edge the side makes a slight but abrupt bend inwards. Above
this is an ornamental scalloped band of incised hatching. L. 75 mm.

Fio. 8. Pottery fragment ; part of rim of vessel, reddish pottery ; 88 mm.
below edge the side bends abruptly inwards ; above are two horizontal bands of
incised lines, and between them a dotted meander, also incised. L. 88 mm.

Fia. 9. Pottery fragment ; part of rim of vessel, coarse grey pottery ; 87 mm.
from edge the side bends abruptly inwards ; above this are three horizontal bands
of indsed lines, the two lower connected by a similar vertical band. L. 74 mm.

Fio. 5 (text). Pottery fragment ; part of side of vessel, black pottery with
reddish sur&ces ; the extmor is ornamented with horizontal^ bands of deeper red.
L. 88 mm.

G 5 '

1, Spearhead Ihalf size), 2, Axebkde (half size). 3, 4, 5, Carved Shells (full size), 6, 7, Flake (half size).

kL.

^ — * j:^ ^

KHJtATUM
I'l.il.i VIII. FiK. a, yo> (liulf si/o) read (one-fourth)

tj 5 '

1, Spearhead (half size). 2, Axe-blade (half size). 3, 4, 5, Carved Sheila (full $he), C>, 7, Flake (half eize).

Plate IX

n 7

1, 2, Adze-blades (retUicetl by J). Z, 4, 6, CloUieada (slightly reduced). 5, Flake (full size),

7, Pea tie and M^jrtar (reduced to J).

Fraynu^nt'* of Pottery

Plate XII

Fragments of Pottt!rj»

ERRATUM
Plate VIII, Fig. 2, fw (half size) rtai (one-fourth)

1, Spearbead (half she), 2, Axe-blade (half size). 3, 4, 5, Carved Sheik (full size). 6, 7, Flake (half eize).

THE ORIGIN OF EXOGAMY

By N0RTHCX)TE W. THOMAS, M.A.

Whatever may have been the case originally, it is probably
true that at the present day marriage prohibitions of some sort
prevail everywhere, and controversies as to the origin of these
regulations have been, and still are, frequent in the anthropological
world. There are, as I have pointed out elsewhere, broadly speaking,
three kinds of considerations which serve as a basis for marriage
prohibitions: (a) in the first place the rules may be based on
kinship organizations such as phratries, totem-kins, and the like ;
where marriage within the group of kin is prohibited, exogamy
is said to prevail ; and the corresponding term, endogamy, implies
that there is a certain body of kin outside which marriage is not
permitted. (6) In the second place locality may be taken as the
basis of the marriage regulations and the rule may run that no
one may marry a village-mate, or one of his own local group.
{c) Finally, marriage regulations may be based on considerations
of consanguinity (blood-relationship) or afl^ity (relationship by
marriage), or both. In civilized communities these prohibited de-
grees are the only kind of marriage regulation in force. Not
infrequently more than one of these kinds of regulation is operative,
and we may have, for example, prohibition of marriage between
those of the same kinship group and, at the same time, prohibition
of mai-riage between those who, though of different kins, are closely
related by blood* It may also happen that over and above regula-
tions based on considerations of kinship or consanguinity, there
is a regulation or, at any rate, a custom, requiring a man to marry
a wife of his own tribe or caste ; to this the name of endogamy
has been often given. As there is a risk of grave confusion unless
this term is qualified in some way, so as to distinguish kin endogamy
from tribal, caste, or local endogamy, it might be well to abolish
the term in the latter sense and substitute for it the word homoio-
gamy ; this and its correlative heterogamy can be qualified by such
ac^eetives as local, tribal, facultative, &c., to express the various

S44

THE ORIGIN OF EXOGAMY

grades ; what is commonly called local exogamy would then be
termed compulsoiy local heterogamy.

Heterogamy has frequently, if not invariably, been evolved
from exogamy, when for some reason the kins have become segre-
gated. But there is no connexion or, at least, no obvious connexion
between prohibited degrees and other kinds of marriage regula-
tions.

Whether it be due to a defective terminology or to some other
eause it has not infrequently happened that students of primitive
sociology, in putting forward theories of the origin of marriage
prohibitions, have not distinguished between the origin of exogamy
and the origin of prohibited degrees ; where the distinction has
been drawn, it has generally been assumed that exogamy is the
earlier development and that regulations based on consanguinity
have arisen as the human race became more enlightened or was
impelled by jealousy to impose limitations of some sort.

Theories of the origin of exogamy or of prohibited degrees,
or of both, have been put forward by J. F. McLennan, Lewis Morgan,
Westermarck, A. E* Crawley, J. J. Atkinson and Andrew Lang,
E, Durkheim and others ; and we may classify the hypotheses into
two main groups, distinguishable as ethical and non-ethicah The
former assume that a conscious desire for reformation of evil
practices developed at some time or another, the latter that the ethical
element in marriage regulations is secondary and that they are
due to superstition or instinct, to less moral causes like female
infanticide, or to more moral movements like the desire to avoid
jealousy within the local group or circle of hearth'mates.

The instinct on which Messra, Crawley^ and Westermarck*
base their theories is the repulsion that exists, or is supposed to
exist, between hearth-mates and those who have been brought up
in close intercourse; Dr. Westermarck also argues that the marriage
of near kin would have evil results, so that those who did not
practise it or who practised it less than their neighbours would
be at an advantage in the battle of life and their stocks ^vould tend
to survive, while the others would tend to die out ; in this way, he
thinks, an instinct against the intermarriage of near kin might
be developed,^

It must be observed in the first place that though marriage

* M^stk Hose, p. 222.

Ibid., p. 352.

Human Marriott, p. 320.

THE ORIGIN OF EXOGAMY

345

regulations of some sort seem to be universal, there is hardly
a single relationship which has been universally held to form a bar
to marriage/ and though it is possible that these are secondary
developments, it remains to be proved that the instinct postulated
by the writers in question is a vera causa ; for, as Mr. Lang ^ has
pointed out, if there is one thing more than another which should
promote incest (on this theory), it is the separation of brothers and
sisters long before puberty, which is such a characteristic feature
of some primitive societies. If such rules of avoidance are in-
tended to prevent incest, the prevention of close intercourse from
youth onwards is, if the theory be correct, precisely the means
which would promote the love it is meant to prevent ; it must not
of course be forgotten that the custom of brother and sister avoid-
ance is life long and does not cease to act as a check at any time
of life. But ex kypothesi the closer the relation and intercourse,
the less danger there was of incest, and we are entitled to ask
why the custom of brother and sister avoidance arose at all if it
removed the greatest safeguard against incest. If hearth-mates
develop an instinct against sexual relations with each other, it would
be unnecessary to separate brother and sister for reasons of sex ;
and it would never occur to any one to projK)se that they should
be sepai ated to provide against non-existent dangers.

This criticism, however, assumes that avoidance is a custom of
human origin consciously instituted with a definite purpose ; and
we are hai^dly entitled to assume this. It is true that the facts
cited by Dr. Westermarck^ as to separation of the sexes among
some of the higher apes— the young males are carried by the father,
the young females by the mother — are hardly sufficient to allow
us to formulate any hypothesis as to the existence in ape society
of anything like a custom of avoidance, or of its germ ; but the
possibility must be borne in mind, especially if it can be shown
that avoidance is known in other orders of mammalia. It need
hardly be said that inquiries into the social organization either of
the primates or of other mammals are at best calculated to aid
us only indirectly in the search for human origins; so far as we
know, there is no near ancestor of man now Hving upon the earth ;
the anthropoids and other apes are merely his cousins, and though
in the main the chimpanzee is zoologically his nearest neighbour,

* Human Marriage, pp. 290 sq. ; Ann6e Sociologique. i. 38.

* Social Origins, p* 240. ^ Human Marriage^ p* IS-

346

THE OKIGIN OF EXOGAMY

more than one other species come nearer to man in respect of
single characteristics. Even were it otherwise and were one species
preeminent in respect of every single point of resemblance to man,
it would still be difficult or impossible to base any cogent argument
on the social habits of apes; for, even with our present limited
knowledge, we can say that some divergence in respect of type of
society is found among the anthropoids; but as the Simiidae are
more closely related inter se than any species is to man, it naturally
follows that the differential evolution which is in evidence within
the family of the Simiidae would be not less but more strongly
marked when we come to compare the Simiidae and the Hominidae.
If therefore we examine the social habits of mammals in order to
throw light on human sociology, we search for analogies and sugges-
tions, not proofs.

To take only a single point, no amount of argument will settle
whether primitive man was or was not jealous. Jealousy is very
marked among many of the higher mammals and it is highly
probable that primitive man was so too ; but even if a more extended
knowledge of ape psychology shows that it is absent among them,
we need hardly regard the fact as conclusive evidence that man
was not jealous in his early stages. The Oneida community is
hardly the modem representative of the primitive human group.

There are numerous types of animal societies, and the relation
between the social type and the form of sexual relations is some-
times very marked, sometimes to all appearances entirely absent.
There may be a band of females and young with an old male as
its permanent head ; or the old males may live apart from the
females for the greater part of the year. The community may
be made up of individuals of all ages and both sexes, or there
may be different groups for different ages for the two sexes, or
for females with and without young ones.

Among the anthropoids the gorilla seems to Hve in families,
consisting of one adult pair and the young ones, the latter being
eventually driven away to form small bands which break up as
they in turn form family groups,* The old Orang Outan, on the
other hand, is solitary, and the remainder live in small groups ; *
while the chimpanzee, according to some accounts, lives in quite

' Human Marriage, p. 13 ; Bu ChaiOu, Vo^a^es (F&fiB, 186S), p. 892,
' Human Marriagt^ p. 13.

THE ORIGIN OF EXOGAMY

347

small groups directed by an old male.^ But in most cases we are
quite in the dark as to the character of the sexual relations of the
apes, whether they are monogamous, polygynous, or polygamous
I in their natural state. In fact it is only in the case of domesticated
or semi-domesticated animals such as horses, camels, and cattle,
[wild or otherwise, that we reaOy know much about these questions,
unless, as is the case with many carnivora, the animals pair and
are, at any rate for a time, monogamous.

As a result of his observations of wild or semi-wild cattle in

[New Caledonia a theory of human social origins was put forward

[by the late Mr. J, J. Atkinson,^ based on the idea that there was

ill the pre- or proto-human condition a herd composed of one old

I male with adult females and young of both sexes. In process of

time, so Mr, Atkinson conceived, the young males, originally driven

out from the herd just as the young bulls are expelled by their

sire, were permitted to remain, greatly to the advantage of the

herd for purposes of defence. The sire at the outset had rights

1 over all the females born in the horde ; consequently it follows that

if, as Mr. Atkinson seems to suggest, the young male eventually

: succeeded to his father's harem, he had rights over his sisters as

[well as over the other women (p< 242) ; tliis, however, conflicts with

' the primal law of brother and sister avoidance, for whose existence

Mr, Atkinson suggested the explanation that only on condition of

such avoidance would a young male be pernutted by his sire to

remain in the horde.

Possibly the passage must be read in another sense ; for the

comer stone of Mr. Atkinson's theory is that the young males

began to introduce females from outside, that the old male was

induced to recognize the exclusive rights of the young males over

these introduced wives, and that when the sire died, so far from

his wives passing to the young males of the group, outside suitors

were found for them who changed their residence and took up

their abode with their wives* As a further development, so Mn

Atkinson supposed, sexual relations came to be forbidden between

fathers and daughters and there came into existence a general

prohibition to marry a group mate.

I Mr. Atkinson's case did not rest entirely on zoological evidence,

I for he argued that, besides being a thinkable hypothesis of human

I social origins, his tlieory gave a more satisfactory explanation of

^^_ * Human Marriagej p. 18. ' Social OriginSf pp. 210-94,

9ft9

THE ORIGIN OF EXOGAMY

customs of avoidance than those offered (for the case of mother-
in-law and son-in-law) by Iiord Avebnry, who attributes it to enmity
caused by the custom of abducting the bride, and by Dr. Tylor,
who stiggests that an outsider had to be ' cut ', at least tiU the birth
of a child had, as it were, legalized his marriaga' Mr. Atkinson
held tliat customs of avoidance — ^the rule that certain relatives mu^
not speak to, or even see, each other — ^are in the main due to jealousy,
and are explicable by the successive steps which he supposes to
have led from the primaeval horde to the savage tribe of the present
day. Thus, the avoidance of father-in-law and daughter-in-law dates
back to the time when the young males first began to introduce
into the group females whom they had captured outside and over
whom they claimed exclusive rights ; the old male having previously
claimed rights over all adult females vrithin the group, enmity
between father and son was only avoided by the adoption of a
device which put the latter's wife out of reach of the sire. Mr.
Atkinson further supposed that alien sons-in-law in time began
to take the daughters to wife before the death of the old male;
here again avoidance between father-in-law and son-in-law provided
a means of smoothing over difficulties, for the latter would be a
usurper of the former s rights and therefore at enmity with him.
In like manner the father-in-law's jealousy compelled the incoming
son-in-law to avoid his mother-in-law, the wife of the old male*

One criticism of this theory of the origin of avoidances at once
suggests itself; if avoidance was a means of diminishing friction
or preventing jealousy, and if father-in-law and son-in-law actually
avoid one another because the latter infringed upon the rights of
the former, it is at least very singular that practically no trace of
father-daughter avoidance is found. One would have imagined that
it was no less needed for the protection of the rights of the alien
son-in-law than the avoidance of his mother-in-law by the intruder
was needed for the protection of his father-in-law's rights. Mr.
Atkinson supposed that father-daughter marriage disappeared little
by little, and so gradually that it left no traces in custom. Now
the relation of the daughters of the young males to the sire of the
latter is a dark point in Mr. Atkinson's theory ; but even if we deal
with the position only of the daughters of the old male himself,
it is clear that the introduction of the alien son-in-law in his life-
time must have speedily broken down any custom of father-daughter

' J. A. /., xviu. 245-6a

THE OBIGIN OF EXOGAMY

349

marriage, and set up 80 strong a body of young males within the
group with interests opposed to those of the old male as to render
the position of the latter untenable in a very short time. Hence
Mx', Atkinson s theory of the alow disappearance of father-daughter
marriage is hardly likely.

Again, except as a part of systematized avoidances to which
the whole of a group and not the individual only are subjected,
it seems probable that there is no hindrance to the meeting of
mother and son, or at any rate not more than to the social inter-
course of father and daughter. Now Mr. Atkinson does not suppose
that mother and eon marriage was ever practised, and the absence
of avoidance between both male and female parents and their
children of opposite sexes is, prima facie, a ground for supposing
that if mother and son marriage was unknown, so too was father
and daughter marriage.

We may therefore ask on what grounds Mr. Atkinson made
sexual relations of near kin a part of his theory. He imphes that
father-daughter unions are found among the half-wild cattle of
New Caledonia, though he nowhere explicitly states it as a fact.
It has already been pointed out that zoological evidence supplies
us at best with analogies ; even if therefore it is a fact that such
unions are known among half-domesticated cattle, it is by no means
incimfibent upon us to accept them as an element in pre- or proto-
human social organization. If it can be shown that among some
of the higher mammals such unions are barred, the New Caledonian
evidence is still further discounted. There is not much evidence
on the point, but autliorities report, though possibly only on the
evidence of natives and not from personal observation, that the
Khirgiz stallion drives out his own fillies as soon as they are in
heat, whereupon they run in a straight line up wind till they come
upon another herd. In the same way tlie young stallion avoids his
dam ; as a rule he is driven out by his sire, just as the young bulls
are driven out, but it may happen that he is restored to the herd
at a later period ; in this case it is neceasary to remove^ his dam ;
otherwise he treats her as the fillies are used by the sii^J
Apparently this account relates to the tame or semi-wild horse ;
but Aristotle ^ relates that the young bull camel has an aversion

* Z,/ R, ill 302; cf. Bi-ehni, Tkrlehm\ iii. 18.
^ Aristotle, UiM, Anim. IX» xjL2iY< 1*

360

THE ORIGIN OF EXOGAMY

to connexion with his dam, and we may accept as a fact the exist-
ence of sexual aversion among animals between parents and offspring.

It is of interest to note that the bull camel can be deceived,
if the female is covered up. From this we may perhaps assume
that the tendency of domestication is, in this as in so many other
cases, to obliterate the instincts of the wild state. Whether this is
so or not, it is clearly immaterial whether man at the present day
usually or invariably shows this kind of aversion. It is a legitimate
hyix)thesis that man once shared this instinctive feeling with
other mammals, even if it can be shown that in our own day in
South America and other parts of the world the union of parents
and children is not uncommon or even frequent. This point is
perhaps the most important objection to Mr* Atkinson s premisses^
but it is far from being the only one. He has taken as his model
the form of society familiar to Mm from observation of half-wild
cattle ; but it is by no means apparent that they are the nearest
analogues to proto-human society. Why, for example, must it be
assumed that there was one old male in the original group together
with several adult females and young of both sexes ? Even if we
reject the possibility of groups constituted by monogamous pairs
and their children as affording no scope for social development,
it is far from clear that the original group did not contain several
adult males, either monogamous or polygynous, each supreme over
his own family and suffering no interference with his sexual rights.
The tarpan of Tartary lives in herds organized on this system, and
so does the jiggetai.' Even were no such examples known to us,
the variety of group type among the apes is such that we must feel
great doubt as to the most probable form of the primitive human
family.

In order to limit the field to be examined we may perhaps
exclude from our purview the multiplex monogamous group (com-
posed of monogamous families), not because such a form of society
is a priori improbable, but because, always assuming that father-
daughter marriage is barred, it would not differ in essentials from
the multiplex polygynous group, save that it would be easier for the
unattached young males to abduct a wife or two, though not so
easy as it would be in the simple polygynous group, which is the
alternative form of society to be considered.

Brehm, Tierlebm \ liL 7, 18.

THE ORIGIN OF EXOGAMY

S51

Parailiarity with the habits of cattle perhaps led Mr, Atkinson
to over-estimate the power of the old male in the primitive human
group ; for it is one thing for the old bull or the old stallion to
hold his own against one or more younger competitors and quite
another thing for an animal so inadequately provided with natural
weapons as was primitive man to attempt to hold his own single-
handed against several rivals. The horns of the bull or the hoofs
of the horse may turn the scale in favour of the experienced and
wary duellist in a very few minutes ; but man had no such
formidable weapons, Tlie gorilla may be able to do much against
a human antagonist with fists and teeth ; but it is another matter
when it is a question of several adversaries of but sUghtly inferior
strength and equally resistant skulls. It may well be that the
simple polygynous group was never a form of human society ; or if
it was, it may have been speedily transformed into a multiplex
group for the sake of the greater protection against a common enemy
which is afforded by the presence of several adult males, ready to
join forces, as do the stallions of the large assembUes of horses.
Practically, therefore, we are reduced to considering the multiplex
polygynous group— Mr. Atkinsons second stage — but without father-
daughter unions.

It was a corner-stone of Mr. Atkinson s theory that adelphic
(brother and sister) unions were prevented by the expulsion of the
young males and the appropriation of the young females by their
sire. If, therefore, the latter factor is eliminated we have to face
the possibility that adelphic unions prevailed, as Lewis Morgan
supposed. If both young males and young females were permanently
expelled from the herd — and it must be noted that among some
mammals the exile of the young males is only temporary and
ceases when they have captured a harem of their own — it might
still happen that the males founded a new group of their own,
while the females joined another, perhaps already existing, group.
Let us suppose, however, that the young males hang about the
outskirts of their original group, as do tarpans and othera of the
horse tribe round the herd in which they were bom, and eventually
rejoin it, Wliat follows?

If the several males in the multiplex group are not at liberty
to appropriate each other s female oflFspring, the aversion between
parent and child causing the latter to be driven out, it would be
equally impossible for the young male to bring the exiled female

352

THE ORIGIN OF EXOGAMY

back to the group. Adelphic unions are therefore barred, even if
the young females are not snapped up by another group.

We have seen that in primitive groups at the present day
brother and sister avoidance is widely practised, while parents do
not commonly shun their offspring of opposite sex. What is the
meaning of this rather striking difference in conduct? It seems
a legitimate hypothesis that in the latter case natural aversion
existed at the outset and was sufficient to prevent unions of this
kind, while in the case of brother and sister there was not
necessarily any such natural bar, and legislation, that is to say
what corresponded to it in those days, viz. the public opinion of
the group, or of the older males, was needed to restrain the inter-
marriage of the offspring of the same jiarents. One way of attaining
this might be by instituting a custom of avoidance, the germ of
which is perhaps seen in the gibbon habit of separation of the
young of different sexes. On the other hand we may perhaps
with more probability regard adelphic avoidance as originally
imposed on adults only, and due to the law which forbade sire and
child to be members of the same group, but later extended back-
wards. But it is by no means essential to the theory here put
forward that any kind of avoidance should have been known in
primitive times.

But, as we have already seen, adelphic avoidance is not the
only custom of this kind ; avoidance between relatives by marriage
is even more prominent, if not more widely found. Mn Atkinson
undertook to explain the genesis of these latter customs by
supposing that in the first place alien females were introduced
into the group not for the sire but for the youthful male offspring,
and that this gave rise to avoidance between father-in-law and
daughter-in-law as a protection of the rights of liis sons ; secondly
that outside suitors came into the group for the daughters of the
old male, at first only after the father's death, later as rivals to him
for their hands, and that out of this arose avoidance between
son-in-law and mother-in-law as a protection to the father-in-law's
rights, and between son-in-law and father-in-law in order to diminish
their enmity as rivals.

Now, as Dr. Tylor s statistics have conclusively shown,' so far
as avoidance gives any clue, the original rule was for the husband
to live with the wife ; then came a period when he lived with her

^ J. A. /., xviii. 251,

THE ORIGIN OF EXOGAMY

S53

family for a time and subsequently removed to his own people ;
finally came an age when the wife was taken to her husband's folk
as soon as they became man and wife. Dr, Tylor was careful to
point out that his data did not necessarily give information as
to the primitive condition of mankind ; but whether this is so or
not, the conclusions seem to invalidate Mn Atkinson's argument.
If avoidance as we know it is primitive, his scheme is contradicted
by the evidence as to the order of the stages brought forward by
Dr, Tylor ; if it is late, it is useless to explain the custom as an
outcome of the conditions of the primitive group.

In point of fact avoidance of relatives-in-law is probably late,
for it originated at a time when the husband resided with the wife
and must have been the product of an age when friendly relations
were already established between different communities ; for, as
Dr. Tylor pointed out, it is inconceivable that a man should remove
at marriage into a hostile community, or even into a neutral one ;
the simple expedient of capturing a wife would permit him to
remain among his friends. We are therefore at most concerned
with adelphic avoidance.

To return to the primitive group, we have supposed that there
existed father-daughter aversion, expulsion of the young females,
temporary exile of the young males and their later return with
brides from another group. Aversion between parents and off-
spring supplies an explanation of the origin of two prohibited
degrees directly, and secondarily of the rule against adelphic unions.
But in the expulsion of one set of females and the introduction of
another we have the principle of exogamy ; and if we suppose that
only two communities were within such distance of each other
and that exchange of females was possible or easy, we have at once
the simplest possible form of exogamy, the intermarriage of two and
only two groups. But at this point we are brought face to face
with the problem of the origin of phratries and totem-kins, and of
their mutual relations. Do these primary groups correspond to the
former or to the latter ?

For there are statements, as Mr, Lang points out in the present
volume, to the effect that within the phratry^ in the Urabumia
and certain other Australian tribes, marriage is further limited to
a single sept (totem-kin) ; and if this should be confirmed, we
have to face the possibility that totemic exogamy is the germ and
phratriac exogamy a later accretion.

TTIXlJt A ft

354

THE ORIGIN OP EXOGAMY

As long as we are confronted with verifiable but unverified
statements Buch as those about the Urabunna system of exogamy,
it is futile to carry the discussion further. It is absolutely necessai*y
to know whether toteniic exogamy within the phratry does or does
not exist ; for on this point must turn to a large extent our answer
to the question, Wliich is earlier— totem-kin or phratry ? ; if indeed
exogamy did not for some reason develop in or become associated
with a previously existing group in which we must recognize the
germ of the totem-kin. For if Dr, Frazer's theory of the origin
of totemism is correct— and i-ecent researches among the Arunta
show that conception is at any rat« a vera cama^ accepted by
a totemistic people as their exphination, or rather as their rule
of the origin of totems — we can hardly regard the exogamic law
as anything but a later accretion. In fact it would not be surprising
to find that the original germ, if so we must t^rm conceptional
totemism, developed in one direction into totemism, in another into
a cult of animals or into magic. But it remains to be shown that j
the Arunta creed of the present day is primitive ; it is therefore j
useless to discuss whether this theory of totemism is subversive of
the hypothesis put forward in the present paper.

Peering into the dim past of the human race, we deal with
possibihties rather than probabilities; liistorical data are denied
us, and if the tmly primitive survives at the present day, we have
no touchstone by which to distinguish it from the accretions of
countless ages. It is a truism that in dealing with peoples in the
lower stages of culture wo cannot always distinguish the rude from
the primitive, nor the back-water from the main stream of human
progress. Some of our uncertainty, however, is due to inaccurate or
insufBcient data ; every year civiUzation and European culture snap
some of the links which bind the dark-skinned races of mankind to
their past, and perhaps it is not too much to say that another
hundred years will see the world * civilized ' ; it may be that the
disappearance of primitive culture is inevitable ; but the age whicli
cuts down its representatives can at least garner the ethnographical
hai*vest before it is too late, for arcliaeology alone can never tell
us the story of the human race.

Plate XTV

^i4 #M# <fS^ 'S

'O^.S-:^'

^iip ^^^

FJO. 1

THE SECRET OF THE VERGE WATCH

A STUDY IN SYMBOLISM AND DESIGN

By ARTHUE THOMSON, M.A.

what is known in the trade as a * Vei^e Watch ', there is
an elaborately engraved plate of circular fonn, which supports the
balance-staflf pivot, and protects the balance wheel. Apart from the
beauty of their design, and the richness of their pierced work and
chasing, these * Watch-Cocks \ as they are called, display other
features of interest, which are weU worth studying, Hie form of
the escapement used in the verge watch is the oldest known;
unfortunately we are ignorant of the name of its inventor. That
it was in existence prior to 1513 is proved by the fact that Leonardo
da Vinci makes mention of it, though pocket watches were not
introduced till about 1625. Tliis style of movement, probably
owing to the simplicity of its construction, continued in use in the
conunoner class of watches up till the middle of the nineteenth
century, when it was replaced by other and more effective forms
of escapement.

Having made a considerable collection of these * watch-cocks ',
the sole relics of wornout and broken-up watches of the verge
type, my attention has naturally been directed to their style and
workmanship. They were made by specialists called ' cock makers ',
men of great skiU as craftsmen and designers. Though most of
these watch-cocks display a strong family resemblance, no two
are absolutely alike, each exhibits the personality of its maker;
they are in every sense artistic productions, and as such are well
worth collecting and preserving. Whilst, as a rule, the gromidwork
of the design is based on the conventional treatment of flowers
and foliage, there constantly recur among the specimens such devices
as grotesque heads or masks, the figures of birds, baskets, vases,
and the like. In grouping the watch-cocks according to the cha-
racter of their design, the question naturally arose as to wliat
interpretation was to be put on the repeated occurrence of the

A a 2

356

THE SECRET OF THE VERGE WATCH

above devices. The most obvious assumption was, that the recur-
rence of the same symbol pointed to the specimens being the product
of the same school of craftsmen, where probably the apprentices
copied and repeated tlie designs of their masters. No doubt such
an explanation may to some extent account for the persistence
of certain types, of what for the time being we may style ornament,
but it doas not adequately explain why the choice of emblems or
devices should be so restricted and limited. Happening at the
time to be interested in the subject of the * Evil Eye *, the thought
natuiully suggested itself to me, that possibly these devices and
patterns partook of the nature of charms, whether to protect the
watch itself or its owner is an open question. With the object
of putting this suggestion to the test I have made a ciireful examina-
tion of all the specimens in my possession, some hundred and sixty
or tliereabouts, with the following results.

Of the examples in my collection, about 50 per cent, have
a grotesque head engraved upon them. These heads seem to have
no apparent connexion with the general design, and look as if
they were introduced to meet some traditional requirement.
Britten ^ who tliinks that the earlier ' Cocks ' were quite plain, dates
the introduction of these quaint heads on them about the year
1700, At first it seemed possible that these might be intended
to represent the Gtorgon's head, but in no instance have I noticed
the split or protruded tongue, the teeth, though often large and
well pronounced, are never tusked, nor have I ever seen snakes
introduced into the design. Support was given to the view that
it was possibly the head of Bacchus, since in a considerable number
of cases the head was represented crowned with flowers. In this
connexion it is interesting to note that Oscilla or masks repre-
senting the head of Bacchus were frequently hung on fruit trees
to avert the evil eye and preserve their fertility. For our purpose,
however, the identification of the head is unnecessary, since it is
generally admitted that the grotesqueness of the mask is in itself
sufficient to divert the attention of the ' evil eye *. El worthy -
writing on this subject, says: *The step fiom the famous death
dealing visage (the Gorgon*s head) as a protection against the evil
eye it was believed to produce, is but short to that of hideous
faces in general; and hence we find that strange and contorted

^ Old Clocks and Watches and their Makers, Batsford. 1904, p» 627.
' TJte Evil Eye. John Murray, London, 1895, p. 147.

THE SECRET OF THE VERGE WATCH

367

faces or masks were certainly used as objects to attract the evil
eye, and so to absorb its influence and to protect the person wearing
or displaying the mask/ There seems, therefore, adequate justifica-
tion for assigning to this class of amulet the devices represented
in fig, 1. In this series, it is interesting to note that in the
majority of cases the chin is only imperfectly shown, in other
instances the design does not include the mouth, but in all the
wrinkled forehead and staring eyes are well displayed.

About 13 per cent, of my specimens have the form of a vase
incorporated with the design. Of this kind of charm Elworthy ^
writes as follows : ' It was said by Plutarch that when Isis brought
into the world Harpocrates, the posthumous son of Osiris, she wore
an amulet round her neck in the form of a vase, the ** Emblem of
Ma " the goddess of truth* The vase is also a symbol of Osiris.
This vase represented water : hence the vivifying power of nature,
i. e, Osiris the personification of the Nile, which was thus typified
by a vase/ In none of my specimens are the vases represented
as ha\ing handles (see fig. 2) like the Amphorae or Canthari
which are shown in the tablet brought by Dr. Ai^hur Evans from
Taranto, and now deposited in the Ashmolean Museum. In
Dr. Evans's opinion this tablet was employed for stamping the holy
cakes used on various religious occasions, and the objects thereon
represented are undoubtedly such as were used as charms. There
appears, therefore, some reason for the assumption that the device
of the vase on these watch-cocks has a similar significance, though
on the other hand it must be admitted that their appearance
corresponds with the re\4val of Greek ornament wiiich took place
towards the latter end of the eighteenth century.

Next in frequency in my collection (about 9 per cent.) are speci-
mens displaying the figures of birds (see fig. 3). Of these, two are
manifestly swans, the othei^ represent birds of prey, more or less
conventionally treated, but in two examples they are undoubtedly
intended for eagles, and the others, there is little doubt, are repre-
sentations of the same bird. It is surely something more than a
coincidence that those birds, and those only, should be selected for
incorporation in the design. Regarded as they are as sacred, from
their association with Jupiter, we meet with repeated examples of
their employment as charms against the evil eye. Elworthy {he. citj

* Iqc. ciL, p. 125,

868

THE SECRET OF THE VERGE WATCH

p. 326) records the occurrence of the eagle four times in specimens
of the mano-pantea, and also figures an engraved onyx {loc. dt^
p. 131) with a central eye surrounded by figures of an eagle, Jupiter's
head, a thunder-bolt, and a dolphin.

The symbol of abundance^ a cornucopia, has long been used as a
chaiTii against the evil eye. It has been supposed to act by exciting
the envy of the Jettaiore^ so distracting his gaze from the object on
whiehj in the first instance, it was directed ; other representations
of plenty are supposed to act in a similar way. It is therefore
interesting to note that baskets abundantly laden with fruit and
flowers are not unfrequently engraved upon these watch-cocks. Of
these, some examples are shown in fig. 4. It is difficult to account for
the occurrence of such designs on the inside of a watch, unless on
some such supposition as that advanced.

Of other forms of device met with, those displaying a radiating
or star-like pattern are not uncommon. It requires but little stret€h
of the imagination to regard the central core as the smi from which
the rays are repi^sented as emerging. Such forms are common
enough in the ornaments used on horse furniture to ward oflf the
influence of the evil eye. Is it improbable that they are similarly
employed here? (See fig. 5.)

There yet remain a few examples of designs not hitherto
referred to. Tlxese are shown in fig, 6, The first two exhibit the
head of the Dolphin, conventionally treated, concerning which
Elworthy remarks that in Roman times the Dolphin was considered
one of the special charms against the evil eye.

In the next two specimens a representation of a lyre is a
prominent feature in the design. The same device is described by
Elworthy as occurring on a mano-pantea, and curiously enough it is
present in the terra-cotta mould presei-ved in the Ashmolean Museum,
which, according to Dn Evans, was used for stamping holy cakes.

Of the remaining examples in the figure, one displays what I
take to be the sacred heart, whilst the other, the only example
I have seen, presents a form, whether intentional or accidental,
reminiscent of the crescent and the disk.

The rest of the specimens, amounting to about 12 per cent, do
not present any features which call for special attention, the pattern
being simply that of a foliated design without any additional device.

Such are the facts. In the course of the paper I have shown
pretty clearly wliich way my own conclusions have been drifting,

THE SECRET OF THE VERGE WATCH

369

but one must hesitate before committing oneself to a definite
theory.

To siunmarize the results, it is, to say the least, extremely odd
that so elaborate and artistic a piece of workmanship should be
hidden away in the interior of a watch, when a simple plate would,
for structural pui-poses, have been just as eflfective. Further, it is
remarkable tliat the devices employed are those which are ahnost
univei*sally employed as charms against the evil eye. Midst the
wide range of objects Avhich lend themselves to decorative treatment,
it is surely strange that the selection should have been limited to
the examples quoted in a collection of over 160 sj>ecimens. It is
instructive, too, to note that we have an analogous ease in the
variety of ornaments used on horse trappings, where there is little
doubt that their selection has been determined by their supposed
efficacy as charms.

Even granting that tlie designs on these watch-cocks were
employed as a means of warding oflf the influence of the e\il eye, it
is hard to account for their occun^nce in watches of English make,
that supposition would necessitate the admission of some alien
influence, most probably Italian. Now the watch and clock making
industry was a product of middle and northern Europe, and spread
southwards. If foreign workmen were employed in England they
would most likely be Swiss, Germans, Flemings, or French. There
were, however, Italian families whose names appear as designers,
and such Frenchmen as Jean Boiu*quet (1723) imd Pierre Bourdon
(1725) published designs for the use of watchmakers wliich display
a mai'ked resemblance to those herein described. At a time when
symbolism in design was much in vogue, those craftsmen no doubt
set the fashion, a fasliion which has since been slavishly followed,
without any clear appreciation of the significance of the devices
employed. This seems a not unreasonable explanation of the matter,
for it is confessedly hard to imagine that the facts already stated
amount to notliing more than mere coincidences.

The clue to the solution of the problem is, therefore, probably to
be found in tracing some Italian influence among the early designers,
or possibly the requirements of the Italian market may have induced
our watchmakers to provide their foreign customers not only with
a reliable time-keeper but also with a sure protection against the
baneful influence of the evil eye.

L-*AR, OR THE TRANSFERENCE OF
CONDITIONAL CURSES IN MOROCCO

By EDWARD WESTERMARCK, Ph,D,

The term l'^\ir is applied by the Moors to a compulsory relation
of a peculiar kind in wliich one person stands to another. The
common expression, Ana f^dr alldh u *drak^ * I am in God's 'dr and
your \irj' implies that a man is bound to help me, or, generally, to
grant my request whatever it may be, as also that if he does not do
so his own welfare is at stake. The phrase, * In God s *drj' only
serves to give solemnity to the appeal : ' I am under the protection
of God, and for his sake you are obliged to help me/ But the word
l-'dr is also used to denote the act by means of which a pei'son places
hknself in the said relationship to another. Had Vdr \lHk^ ' Tliis is
*dr on you,* is the phrase in common use when an act of this kind is
perfonned. If the person so appealed to is unwilling to grant the
request, he answers, Had I'dr yihruz fik^ * May this *dr recoil upon
you/

The constraining character of l*dr is due to the fact that it
impUes the transference of a conditional cui^se : — If you do not
do what I wish you to do, then may you die, or may your children
die, or may some other evil happen to you. That l^'dr implicitly
contains a conditional cui'se is expi-essly stated by the people
themselves, although in some cases this notion may be somewhat
vague, or possibly have almost faded away.

Externally the custom of l-dr presents such a variety of foims
that, without the aid of a connnon term, it might be difficidt to
recognize them all as expressions of one and the same idea. The
only feature which all these acts have in common is that they serv^e
as outward conductors of conditional cui-ses. A common method of
performing \ir upon a person is to estabUsh material contact with
liim by means of a bodily griisp* He may touch the person whom
he invokes with his turban or with a fold of his dress, or he may

362

L-'AR, OR THE TRANSFERENCE OF

grasp with his hand either that person himself or his child or the
horse which he is riding. Even by going to a horse in a stable and
saying, Ana /*ar l-atui^ * I am in the Uir of the horse,' a person may
place himself mider the protection of its owner ; thus people often
take refuge in the Sultanas stable. You may also make ^dr upon
a person by taking his son in your arms and giving him to his
father, saying, 'This is 'dr for you/ Another form of '«r-making
is to take some food to the person invoked. If he cannot or will
not grant the petitioner s request, he refuses to accept the food ; but
if he accepts it he is bound to do what is asked of him.

Among the Ulad Bu 'Aziz in Dukkdla and other Arab tribes of
the plains, if a refugee enters another person's tent or only takes
hold of the tent-pole, at the entrance of the tent, saying, Ana fi *dr
Mh u \irAk, or, Ana zaug ji uVidhUk^ ' I am in God's ^dr and your ^wr^
or, ' I am seeking refuge with your children/ then the owner of the
tent is obliged to assist him, at least by acting as a mediator between
him and his pursuer. A similar nile prevails among those Arabic-
and Berber-speaking tribes who live in houses. Among the Shluh
(Berbei-s) of Glawi, in the Gre^t Atlas, the i*efugee invokes the owner
of the house in the following words : — Zugg guftis nrdbh^i dttinfiik
dftan^ * I am seeking refuge in God's and your hands, O So-and*so/
In such circumstances custom also requires that the pursuer shall
not try to take the refugee out by force, but apply to the owner of
the tent or house ; nay, even if he is pursued by Ms governor, the
soldiers first make representations to the protector to give up the
refugee. Among the UMd Bu * Aziz, if a sheikh or governor wants
to extort money from one of his subjects and the latter takes refuge
in the tent of some important person who happens to be away from
home, the wife of his absent host takes off her belt {l-fizdm) and
gives it to him. The refiigee then goes with it to the sheikh or
governor ; it is 'dr on the part of the wife, and one-half of the claim
will be remitted in consequence. The governor may also himself
induce a refugee to leave the tent where he has fled by sending him
his own rosary as a guarantee that he will not treat liim badly. In
this instance the pursued person gains power over his pursuer by
possessing a thing which belongs to the latter. Tlie following case
occurred in a village among the Andjra mountains in Northern
Morocco, where I was staying for some months. A man who had
committed murder came once running to the village, pursued by the
relatives of his victim. He found on a field some women belonging

CONDITIONAL CURSES IN MOROCCO

363

to the family of my host^ a highly respected shereef, and cried out,
Ana mzaug falldk u Ji&senf, ' I am seeking refuge with God and the
shereef/ Without touching them he lay down on the ground, and
they covered him with some of their clothes. The pursuers were
thereby prevented from carrying out their intended revenge, and
turned back to their \nllage. In Andjra it also happens that a
person who has been subject to unjust punishment goes to a place
where some men are gathered practising rifle-shooting, bows down
with his hands on his back, kisses the ground in four directions, and
says, Ana f^ar aUdh u \irkum ya mivMin I-mkdhdl infu fia Ulleh Siktm
*a! alldh u \iUkum mm medlmn^ ' I am in the 'dr of God and your *ar,
O masters of the guns, attend to me for the sake of God, I make an
accusation to God and to you, I am innocent/ He remains bowed
down with the right hand clenched behind his back and the left
hand clasped round its wrist till the sportsmen have made hfSi^ha
Unllliulm^ ' the reversed fit'ha^' that is, till they, with the palms of
their hands turned downwards, imprecate evils upon his enemy.
Wlien this is done, the chief of the sportsmen opens the hands of
the supplicant, and all persons present clap their hands.

At Amzmiz, in the Great Atlas, I was told that if a man has
committed a rape on another man's wife, and the offended husband
is not strong enough to avenge himself, he makes a hole in a kettle,
hangs the kettle round his neck, and goes about in this manner
asking people to help him. This was said to be 'dr of a very
compulsory kind. Among the UMd Bu 'Aziz a man whose wife
has been seduced by another person or whose property has been
interfered with, may go to his governor with a piece of his tent-cloth
over his head. This is *dr upon the governor, who is now compelled
to help him. Or instead of covering his head with the tent-cloth he
may paint his face with cow-dung, or he may shave his hair, leaving
only a lock (l-gr^n) on each side of tlie top of the head and a fringe
{1-giipm) over the forehead. Among other Arabs of the plains an
injured husband who is too weak to avenge the infringement of his
rights leaves seven tufts of hair on his head and goes to another
tribe to ask for help. In these cases the conditional curse is
obviously supposed to be seated in the kettle, or the tent-cloth, or
the cow-dung, or the locks or tufts of hair, and from there to be
transferred to the person or persons who are invoked, A similar
idea undoubtedly underlies the custom of making a vow of blood-
revenge by letting the hair grow until the vow has been fulfilled.

864

l-*Ar, or the transference of

Ttis is 'dr made by a person upon himself: the conditional self-
imprecation or oath clings to his hair and will ftdl upon his head if
he violates it. Speaking of the same custom practised by the ancient
Arabs, Wellhausen suggests that the hair was allowed to grow for
the purpose of being sacrificed after the fulfilment of the vowj But
this explanation, far-fetched by itself, is in my opinion disproved by
the facts to which 1 have just referred.

Sometimes i-'dr consists in making a heap of stones. Both
among the Aiabs of Uldd Bu 'Aziz in DukkAla and among the
Andjra mountaineers I found the following custom prevalent. If
two men agree to meet at a certain place at night for the purpose of
going out together to rab and one of them fails to appear, the other
man makes a cairn at the appointed place and takes the faithless
comrade to it the next morning. The latter is then obliged to give
him a nice entertainment. The cairn may on the one hand serve
as a proof that its maker had kept his engagement ; but its chief
objecl is without doubt to compel the other person to compensate
him by a feast, the cairn being *dr upon him. In this case, as in all
kinds of l-dr^ the cui*se is conditional ; but unconditional curses are
also frequently embodied in heaps of stones. Thus in the north of
Morocco, if a muleteer buys a new mule, the other muleteers of the
place ask him to give them an entertainment, and if he refuses they
make a cairn asking God to send misfortune on the mule which he
had bought. A common practice among scribes is to make a cursing
cairn for a wealthy man whom they have in vain asked for a present.
Tliey make a cairn either outside his house or in some open place,
read over it some passages of the Koran, and, with the palms of
their hands turned downwards, pronounce a curse upon the niggard,
invoking God to deprive him of his wealth or calling down on him
some other misfortune. Among the Berbei-s of Aglu, in the Sus
province, each of the scribes takes a stone from the cairn which they
have made and throws it away, saying, in Shelha, Gikad Hi niitit
takdrk6rtad aflUaslMtit rAhUi gailli tisfUrhdn^ 'As we dispersed this
heap of stones, so may God disperse for him that which makes liim
happy/ The person on whose account the cairn was made is thus
cursed by the scribes ; he is, as the Arabs say, nishdt ttulbd. Cursing
is the essence of the ceremony. The reading of the Koran, again, is
an imitation of the funeral rite ; the scribes say that the cairn is the
tomb of So-and-so. In other instances the scribes plunge a knife
* Wellhausen, Restc arabisdten Hddmtums^ p, 124.

CONDITIONAL CUESES IN MOROCCO

865

into the ground, cover it with white clothing representing the shroud
of a corpse, and recite l-btinhi, just as at an ordinary buriaL Or they
perform the funeral ceremony over seven little stones wliich they
have wrapped up in clothes.

The most powerftil of all methods of making 'rir, however, is to
sacrifice an animal on the threshold of the house or at the entrance
of the tent of the person from whom a benefit is asked. If he steps
over the blood or even only catches a glimpse of it, he is, for his
own sake, obhged to grant the request made by the person who
killed the animaL If he has previously heard that an animal has
been slaughtered outside his dwelling and he is unwilling to do
what is asked of him, he tells his servants to remove the dead body
and to wash away the blood carefully, and in this case, when he has
not seen the blood at all, the danger is much lessened. On the
other hand, if he fidfils the wish of the supplicant, he need not be
afraid of stepping over the blood, as the curse it contains is only
conditional. In some parts of Morocco^ at least, it is the rule that if
a person who has been thus appealed to is unable to give the assist-
ance required, he is obliged to provide another animal to be killed
as 'dr for somebody else. When an animal is killed as ^dr^ the
ordinary phrase bismiUdh^ ' In the name of God,' is not uttered ; and
it must not be eaten either by the sacrificer or by the person on
whose account it is sacrificed. It is eaten only by the poor.

The great efficacy ascribed to this form of l-*dr i>s entirely due to
the blood. As I have pointed out elsewhere,^ the efficacy of a curse
depends not only upon the potency which it possesses from the
beginning, but also upon the vehicle by which it is conducted — just
as the strength of an electric shock depends not only on the original
intensity of the current, but also on the condition of the conductor.
And of all conductors of curses none is considered more efficient
than blood. The reason for this is that blood is supposed to contain
supernatural enei^, and it is a general law of magic that a medium
endowed with supernatural energy gives particular potency to any
curse with which it is loaded. In Morocco blood which has been
shed is always supposed to contain zmln^ or evil spirits ; there is
misfortune in it — l-b&s yimii nia ddem. The most a%vful of all *dr-
sacritices is the so-called t**arkiha or Vargiha, The sinews of the knee-
joints {l-ardkah^ or Varligeh) of a bullock s hind-legs are cut, and in this
state the animal is taken to him or those to >vhose loss the *ar-sacrifice
* Origin and Bcieloptnent of the Moral Ideas, i. 580,

I/AB, OB THE

Tbmm
to be to KttB che

m Ml appeal to
novMi for cutting
anfiTOil the ap-

ii to be made. T
one tribe mrokes
the Soltan c^- to Mfi
tbe smewB of the bolloek
pouBim of a flippiuBil.

As a meam liy which one petaoa can compel another to comply
with his wiabesy l*'dr oataaUj fhym a rerj important part in
social life of the people^ It h resetted to for a Tariety of
to obtain pardon fnnn the goremmeat ; or to indnee the relatires of
a person wha haa been killed to abstain from taking revenge ; or to
decure mwtmrm against an enemy or m^iiadon in the case of
trouble. A woman once wanted to put 'dr upon me outside m j tent
in order to compel me to give a new cloak to her little son. L-'m-
k thus a great boon to weak and helpless people, criminals, and
strangeEBL The &ct that a person is in the *dr of the owner of a
dwelling as soon as he has entered or even touched it, largely
explains the stringent claims of hospitality; for, by being in his
dwelling, the stranger is in close contact with the host and his
beloi^ings and is consequently able to transmit to them any evil
wishes he pleases. In the Great Atlas mountains a Jew who settles
down in a Berber village always places himself under the protection
of some powerful man by putting 'dr upon him. L'dr makes
travelling possible in districts which otherwise would be in^
accessible even to Mohammedan strangers. For instance, if a man
belonging to the Arabic-speaking Beni Ah'sen tribe goes to visit the
neighbouring Berber tribe, Z6mmur, or a Zemmur man goes to visit
Bdni Ah'sen, the visitor must secure protection {l'^me^rag) from some
member of the other tribe. In case the protector should desert his
protege, the injured party, or, if he has been killed, his relativeS|
make a picture of the faithless man and take it about from marked
to market tolling the people which person it represents and at the same
time cursing him. This is done for the purpose of compelling him to
pay compensation ; and should he refuse to do so a fight may ensue
between the families of the t%vo parties. Or the injured person
threatens to dig the other's grave in the market place, and if his
threat is of no avail he carries it out, announcing, ^So^and^so h^^
broken his word, this is his grave,' in which case the person in
question is regarded as a dishonoured man. A similar system pre*
vails among the Jbala, or mountaineers of the North, and the Shluh,
or Berbei-s of the South, in districts where the Sultan s rule is merely

;

CONDITIONAL CURSES IN MOROCCO

367

nominal ; the JbMa call such protection zzettat The degree of pro-
tection which /-'dr affords to him who resorts to it depends, of course,
upon the respect in which the protector is held. The protection
given by women is particularly powerful. In various tribes of
Morocco, especially among the Berl>ers and Jbila, a pei-son who takes
refuge vrith a woman by touching her is safe from his pursuer ;
whilst among the Aiabs of the plains this custom is dying out,
probably owing to their subjection under the Sultan's government.
The reason why women are regarded as able to offer an asylum is
obviously the belief in their magic power and the great efficacy of
their curses.'

I have so far dealt with Vdr as practised in the relations between
man and man. But it is equally often resorted to as a means of
putting pressure upon supernatural beings, mun and dead saints,

Tlie mun {djinn} form a special race of beings, created before
Adam, They have no fized forms, but may assume almost any
shape they like. They may be met with anywhere, but certain
places are specially haunted. Some of them are good, others bad,
but the latter hold a much more prominent place in the popular
creed. They are always ready to attack human beings ; hence
various means are used for keeping them at a distaiice or appeasing
them,=^ and one of these means is i-Vin

As every place has its inww, the Moors, whenever they build a
house or pitch a new tent or dig a well, make a sacrifice to the spirit
owners of the place, mwMin I-mMn ; and this sacrifice is generally
described as *ar, although the meat of the slaughtered animal is eaten
by the people themselves. The kiDing of the victun takes place at
the entrance of the tent or on the threshold of the new house,
or on the spot where the threshold will be built ; and in many or
most parts of Morocco an animal is slaughtered not only when the
foundation of the house is laid but also when it is ready, or nearly
ready, for occupation. So also when a well is being dug a sacrifice
may be offered not only at the commencement of the work but also
afterwards, when the water first appears, or when the well is ready.
The Berbers of Aglu slaughter a black goat at the place where the
well is to be dug, saying some words hke these; — Ngdrs filTmm
alemluk wamanml afillagtdr/mm fidrko fUVaun adagart*drdem krdn

* See my Origin and Developtnent of the Moral Ideas, I 666 sqq*

* See my article, *The Nature of the Arab 6!nn illustrated by the Present
Beliefs af the People of Morocco/ in Joui\ Antfir. Inst,, xjlijl, 252 sqq*

368

I/AR, OR THE TRANSFERENCE OF

tgmisa nafUlag Ukhan^ * We killed for yoii, O masters of this water, may
you make it easy for us, we shall make it easy for you, may you not
cause us any difficulty/ In Andjra a black goat is killed and thrown
into the stream on the spot where a new millstone is going to be
placed ; this, too, is Vir for the mm. Very commonly Vir-sacrifioes
are oflfered to the znun in cases of illness. Sudden disturb-
ances of the health, such as convulsions, epileptic and paralytic
fits, rheumatic or neuralgic pains, are said to be caused by these
spirits ; and a conmnon remedy is to kill a cock on the spot where
the patient is supposed to have been stinack, or to kill a cock over his
head and then take it to a place haunted by hiutu In either case
the sacrifice of the cock is regarded as 'dr.

Of considerable interest is the 'dr made for dead saints. Under
the name of dead saints I include not only deceased men or women
who were supposed to be endowed wath baraka^ hoMness, already
during their hfetime, but also those purely imaginative beings
who have been invented to give an anthropomorphic interpre-
tation of the holiness ascribed to certain places or objects of
nature. Moorish saints may be divided int43 two classes — such as
really exist or have existed in human sshape, and such as have
never existed at all. Language itself indicates a certain confusion
between a holy person and a holy place. The name siyid (pL sdddt^j is
given both to a saint and to the place where a saint has, or is sup-
posed to have, his tomb or where such a person is said to have sat
or camped. It is always marked in some way or another : by a white-
washed house, or a room without a roof, or a low enclosure of stones,
or merely a cairn. Such a place — and Morocco is crowded with
places of this description — is visited by persons who desire to
invoke the saint to whom it is dedicated, with a view to being cured
from some illness, or being blessed with children, or getting a
suitable husband or wife, or receiving help against an enemy, or
deriving some other benefit from the saint. To secure Ms assistance
the visitor puts *dr upon him ; and here again different methods may
be adopted.

The \tr made on a saint may consist in throwing stones on a
cairn connected with the styid. As already said, a cairn may itself
be a siyid, and very frequently cairns are found in the immediate
vicinity of sMdt'. Moreover, there are cairns on the roadside,
especially on the tops of hilLs, at the place where a slyid first be-
comes visible to the traveller. Tlie Arabs of the plains call a cairn of

CONDITIONAL CURSES IN MOROCCO

369

this sort I'kdrkor dyil rrlUfuha^ the JbAla call it iratida (or rrtiidu)
fi^fak. Passers-by throw some stones on it, thereby placing them-
selves under the *rtr or protection of the saint ; in Andjra it is the
custom to throw three stones, which are first kissed by him who
throws them. But a person may also, with the face turned towards
the siyki^ ask for the fulfilment of some special wish, and in this
case, at least in Andjra, the petitioner kisses the cairn. At the
sanctuaries of saints, or at the place from which a siyid becomes
visible, I have frequently, especially among the Shluh of the Great
Atlas, seen small piles consisting of a few stones one put on the top
of another. These piles are *dr on the saint, made by sick persons
or other petitioners, who generally at the same time promise to oflfer
him a sacrifice if their wish is granted. Wlien the i^etitioner finds
that the saint has listened to his request, he fulfils his promise and
knocks down the pile ; whereas, if the saint gives no assistance, the
pile is left as *dr on him.

In all parts of Morocco it is common to tie rags to objects
belonging to a shjid — to the window-rails of a saint's house, to some
tree growing on a saintly spot, or to a stake thrust in the cairn of a
saint, or to the cairn itself. In very many cases at least, the tying of
rags is Vtr upon the saint. In the Great Atlas I visited a place where
the great saint Mulai *Abd-ul-KMer has a heap of stones dedicated to
him, A large number of rags were tied to a pole stuck in the cairn,
and when I asked for an explanation, the answer was that the
petitioner generally fastens a strip of his clothes to the pole mutter-
ing some words like these : — A}ja0rrdmud^ hdyifi migdk urdig adakzdug
ard itakdut tagausdnUj ' O saint, behold ! I promised thee an oflfer-
ing, and I will not release (Uterally * open ') thee until thou attend est
to my business.' If his wish is fulfilled the person goes back to the
place, offers the sacrifice wliich he promised, and unties the knot
which he made. Among the Ulad Bu 'Aziz a petitioner may go to
the tomb of a saint and tie his turban round one of the corners of
the box {ttahiit) under which the saint lies buned, and the turban is
left there for a night as 'dr on the saint. A Berber servant of
mine from Aglu in Sils told me that once when he was in prison he
invoked LllUa Rah'ma Yusf, a great female saint whose tomb is in a
neighlx)uring district, and tied his turban, saying, * I am tying thee,
LdUa R^\h'ma Yusf, and I am not going to open the knot till thou
hast helped me, nor shall I ever invoke thee if thou dost not assist
me/ And the same night his chains were opened by her and he

TTLOK

370

L-'AR, OK THE TRANSFERENCE OF

escaped ; the saint was e\adently frightened by his threat. Very
often petitionei's knot the leaves of some palmetto or the stalks of
wliite broom growing in the vicinity of a siyid. This is practised all
over the country, and a common rule is that the tying should be
done with the left hand. At Sidi T<ilha s sanctuary in Andjra seven
knots are made on the leaves of a palmetto ; and the same is the
case at the tomb of the patron saint of Tangier, Sidi Muhammed
1-hadj, where the seven knots should be made one after the other
and without difficulty, as it is believed that he or she — this is mostly
done by women — whose hand gets too tired to finish the knots
properly will not have the request granted. So also a native of the
tribe Massa in Sus, when in distress, vnH go to the tomb of Lalla
Rfth'ma Yusf, and knot the leaf of some palmetto growing near her
grave, saying, Ker'fgkem gid aldUa raffma ytisf^ unig addmfsiy gar igiyi
ifmt gtiissMml Ui gilHg, 'I tied thee here, O LAlla R&h*ma Yusf, and
I shall not release thee unless thou releasest me from the toils in
which I am at present.' *

However, the practices of throwing stones at the cairns of
saints, of making piles of stones to saints, of tying pieces of clothes
at sdd(U% and of knotting the leaves of palmettos or white broom, may
have another object besides being *dr on a saint When performed
by persons who are suffering from some illness, the idea of disease-
transference is often conspicuously present in their minds. Outside
the famous Imi ntakkdndut in Hjiha (Southern Morocco), which con-
sists of two huge caves supposed to be inhabited by saintly Inun,
there are innumerable small piles of stones made by visitors
evidently with a view to transferring their diseases to the stones.
Before piling up the stones they rub them against the affected part
of the body, and it is generally assumed that if anybody happens to
overthrow one of these piles he will catch the disease of its maker.
In Andjra, at least, the behef prevails that if a person rids liimself
of a disease by tying some piece of his clothes at a siyid^ the disease
would be transferred to anybody who aftei-wards appropriated the

* These facts have suggested to me an explanation of the Latin word relish,
religion, wliich is probably related to religart\ which means Ho tie*. It ib com-
monly assumed that the relation^jhip between the words implies that in religion
man was supposed to be tied by liis god. But I venture to believe that the con-
nexion between religio find reUgare allows of another and more natural interpretation,
namely, that it was not the man who wa8 tied by the god, but the god who waa
tied by the man, Tho Romans were much more addicted to magic than to true
religion ; they %vanted to compel their gods rather than to be compelled by them.

CONDITIONAL CUESES IN MOROCCO

371

strip. Among the Berbers of Ait Zfel'dn, in Haha, sick people used
to visit a miracle-working wild olive-tree growing in the immediate
neighbourhood of the supposed grave of Sidi Butlila (* my lord the
master of relief '). They there rid themselves of their complaints
by tying a woollen string to one of its branches ; in cases of headache
the patient previously winds the string three times round the top of
liis head, whilst in case of fever he spits on the strings and when ty-
ing it to the tree says, FM{)gik taukinu ayazimmUrad^ ^I left my fever
in thee, O wild olive-tree.' He believes that he may in this manner
transfer liis disease to this particular tree because there is bamkuj
* benign virtue,' in it ; he would not expect to be cured by tying the
string to any ordinary tree. The transference of evil is not looked
upon as a merely * natural ' process ; it can hardly be accomplished
without the aid of magic energy; hence acts calculated to bring
about such transference are performed by contact with some holy
object. The making of knots may serve a similar purpose. Near
Mehdiah, on the Atlantic coast of Morocco, I found at the road-side
bushes of white broom with the tips of their stalks twisted into
knots, and I was told by two shepherd boys who happened to be on
the spot, that when a person is suffering from backache he makes
such a knot with his hands behind liis back. But whilst acts of this
kind imply disease-transference, they may very well at the same time
be *dr for a saint. We should remember tliat l-'dr itself is an act of
transference — the transference of a conditional curse.

A fusion of diflFerent ideas may also be found in some other
methods of making *dr upon saints. If a person goes to a siykl to
invoke the saint's revenge upon an enemy, he sometimes takes with
him some burned com and strews it on the floor of the sanctuary,
saying, Sett*et% ya sMij fltln kimu' Mifet' had SSerJik, * I threw, O saint,
So-and-so as I threw this com/ This is 'dr on the saint, but at the
same time it is an act of symbohc magic* For a similar purpose
supplicants burn tar inside a siyidj or, if the door be closed, outside
the entrance, maintaining that they are thus burning the enemy.
Again, if a person has been falsely accused, he turns over the carpet
in the siyid which he visits, asking God to turn a still greater
accusation over the false accuser. Another method of calling doT-vn
misfortune upon an enemy is to sweep the floor of the siyid with one's
cloak, praying God to sweep the enemy likewise. It may be that
in these acts symbohc magic predominates, but they are all called
/-Vir, and are considered to compel the saint to give his assistance,

B b 2

372

VAR, OR THE TRANSFERENCE OF

As in the case of l-dr made on living men, so also l-'dr made on
departed saints, or at sMdt" generally, very frequently consists of an
animal sacrifice. This *ar-sacrifice is accompanied by a promise to
reward the saint if he grants the request ; the Moors are too shrewd to
give a present before they know that they gain by it The reward
given in fulfilment of such a promise is called 1-ivada. It may itself
consist in a sacrifice, but one totally distinct from that olFered as *dn
Whilst the ar-sacrifice is a means of constraining the saint, the ivcTda-
sacrifice is a genuine gift. An animal which is oflfered as *dr to a saint
is killed without the usual phrase bimiilldhj * In the name of Grod '; and
it may not be eaten, except by poor people. However, if the siyid has
a mktUlddm^ or regular attendant, he tries to induce the petitioner to
give him the victim aUve, so that he may himself kiU it ' in the name
of God ', and thus make it eatable. Then the descendants of the saint
(if he has any) and the mkdddSm himself have no hesitation in eating
the animal, even though it was intended by the visitor as 'dr on the
saint; bismilldh is a holy word which removes the curse or evil
energy inherent in Ir'dr. On the other hand, the animal which is
sacrificed as wadu is always killed ' in the name of God *, and it is
offered for the very purpose of being eaten by the saint's earthly
representatives. Nothing can better show than the Moorish distinc-
tion between l-dr and l^wada how futile it would be to try to explain
every kind of sacrifice by one and the same principle. The dis-
tinction between them is absolutely fundamental. The former is
a threat, the latter is a promised reward.

From the words which are uttered when U*dr is made on a saint,
it is obvious that in such cases, also, Vdr impKes a conditional curse.
When I have asked how it is that a saint, although invoked with
I'dr^ does not grant the request made to him, the answer has been
that the saint does all that lie can, but that he is not all-powerful
and the failure is due to the fact that God does not Usten to his
prayer. But it also occurs that a person who has in vain made *dr
on a saint goes to another sinid to complain of him. There is a
general beMef that saints do not help unless \ir is made on them — an
idea which is not very flattering to their character.

As a person is placed under another person's protection by
coming in contact with his dwelling, so <%lso anybody who takes
refuge to a siyid is in the \ir of the saint. As a refugee may, by his
curse, compel a hving man, so he may in a simihxr manner constrain
a saint as soon as he has entered his sanctuary. The right of sane*

CONDITIONAL CURSES IN MOROCCO

878

tuary is regarded as very sacred in Morocco^ especiaUy in those
parts of the country where the Sultan's government has no power.
To violate it is an outrage which the saint is sure to punish. In a
village in Garbiya I saw a madman whose insanity was attributed to
the fact that he once had forcibly removed a fugitive from a saint^s
tomb ; and of the late Grand- Vizier it is said that he was killed by
two powerful saints of Dukkala, on whose refugees he had laid
violent hands. Even the descendants of the saint or his mkidddm
can only by persuasion and by promising to mediate between the
suppliant and his pui'suer induce the former to leave the place.
This shows how anxious the saint is to protect his refugees. He
must protect them because they are in his *at\

Closely connected with l-dr is l*dhM^ or covenant. Whilst Par
is one-sided, l-ahM is mutual ; both parties transfer conditional
curses to one another. And here again the transference requires
a material conductor. In times of rebellion chiefs exchange their
cloaks {ssldMm) or turbans, and it is believed that if any of them
should break their covenant he would be punished with some grave
misfortune. This practice I found prevalent both among the Arabs
of the plains and the Berbers (Briber) of Central Morocco, Among
the Ulad Bu *Azi2 it is a common custom for persons who wish to
be reconciled after a quarrel to go to a holy man and in his presence
join their right hands so that the fingers of the one go between the
fingers of the other, after which the saint throws his cloak over the
united hands, saying, Edd VdhM blmltkum, ' TliLs is *dhM between
you.' Or they may in a similar manner join their hands at a saint's
tomb over the head of the box under which the saint is buried, or
they may perform the same ceremony simply in the presence of
some friends. In either case the joining of hands is usually accom-
panied by a common meal, and frequently the hands are joined over
' the dish after eating. When joining the hands they say, Hdda
'dhidUdh bini u hinik^ *Tlns is 'dhM of God betw^een me and you.'
And if a person who has thus made a compact with another is after-
w^ards guilty of a breach of faith, it is said of him, Rdhb^i u tCdm
ihdUd% * God and the food will repay (him) * ; in other words, the con-
ditional curse embodied in the food which he ate will be realized,
Similar fonnsof h'dMid are practised among the Berbers of the Great
Atlas and Siis ; and all over Morocco the usual method of sealing a
compact of friendship is by eating together, especially at the tomb of
some saint. The sacredness of the place adds to the efficacy of the

374

COKDITIONAL CUESES IN MOROCCO

imprecation, but its vehicle, the real pimisher, is the eaten food|
because it contains a conditional curse. The UIM Bu 'Aziz say that
it is more important to be kind to a neighbour than to a relative,
because neighbours so often take their meals together.

The *dMd of the Moors helps us to understand the covenant
sacrifice of the ancient Semites, The only difiFerence between them
seems to be that the former is a method of establishing a compact
between men and men, whilst the latter estabUshed a compact
between men and their God. Tliat the idea of transferring a curse
by means of a sacrifice was familiar to the ancient Arabs is obvious
from the custom of throwing the hair of a sacrificial victim on a
holy tree as a curse ; * and in the covenant sacrifice the curse was
mutual. They sealed covenants by applying sacrificial blood to the
sacred stone representing the deity, and the worshippers, on their
part, dipping their hands in it.- In the covenant ceremony at Mount
Sinai half of the blood of the sacrificed oxen was sprinkled on the
altar and the other half on the people. ' Among the Hebrews, as in
Morocco, covenants were further made by the parties eating together ;
and this was also the case with covenants made mth the deity.* In
the light of the Moorish \ihad the meaning of this sacrificial meal
seems clear. There is no foundation for the theory laid down by
Robertson Smith and his followers that it is a survival of a previous
custom, according to which the god— that is, the totem god — himself
was eaten, and that it was a sacrament in which the wholb kin, the
god with his clansmen, united, and in partaking of which each
member renewed liis union with the god and with the rest of the
clan. The object of a sacrificial meal may be to transfer blessings
to worshippers ; for, by being oflfered to a god, the victim becomes
instinct with beneficial supernatural energy, which by eating is
transferred to those who partake of the meal.'^ But when the
purpose of the sacrificial meal is to establish a covenant, the idea
underlying it is not to transfer blessings to the worshippers, but to
transfer conditional curses both to the worshippers and to their god.

' WeHhatisen, op. ciL^ p* 124. So little has the true import of such sacrifioos
l>een UDderstood, that Wellhausen represents the one in question as a gift to the
deity,

' Robertson Smith, Religion 0/ ilte Semites, p.SH; Wellhausen, op. ctl, p» 125v

* ExoduSy xxiv. 4 sqq.

* Robertson Smith, cjp. cit^ p. 271 ; Wellhauaen, op, cit., p, 124,

* See my Origin and Devetqpmeni of the Moral Ideas, i/445 sq.

A BIBLIOGRAPHY OF
EDWARD BURNETT TYLOR

FROM 1861 TO 1907

COMPILED BY BARBAR/V W, FBEIRE-MAHRECO
LADY MARGARET HALL, OXFORD

The following abbreviationg are employed : —

Acad. — Academy.

B. = Bodleian Library Catalogue.

B.M. = British Muaeum Catalogue.

Brit, Abb, R, = Report of the British Aaeocia-
tion for the Advancement of Science

Fortti- Rev, = Fortnightly Review,

Joum. = JournaL

Joum. Anth, Inst. = Journal of the Anthro-
pological Institute.

Mag. = Magazine.

Pop, Sci Mo* = Popular Science Monthly.
Proc, Royal Inat. = ProceedingB of the Royal

institution.
Quart. Rev, = Quarterly Review.
Rev. = Review.
Trans. = Transactions.
Trans. Ethn. Soc- = TraDsactions of the

Ethnological Society.
Univ. Gaz. = Oxford University Gazette.

iseL (1) Anahuac : O R Mexico and the Mexicans. Ancient amb MoBgRy,

London: Longman, Green, LoTigman, and Roberts. 186L 8vo.

xi + 344. Coloured plates ; map ; woodcuts.
B. [208, a. 318]. B.M. [2374. e. 17],
1662. (2) Beinarks on Busobmanii's ' Hesearches m North American

Philology*: Beinew of Busehmaons Spuren der Aztekischen

Sprache im Nordlichen Mexico ; 1859, <frc.
TraoB. Ethn. Soc, 2: 130-136.

(Diitrihution of trihes— Relation of Sonora family to Aztec— Permanence of
American languages.)
* 166S. (3) WUd Men and Beast-Children.
Anthropological Rev., 1 : 21-32,

(Wild children in Gerraany, India^ Poland, Ac.; stories of beaet- children;
simultaneous births of animak and children.)
1864, (4) Biscovery of a Celtic Kitohen-Heftue^Heap at HormaDby in Cleve-

lanci, (Anon.)
Gentleman s Mag,, 16: 162-167,
• 1866. (5) Reseahches into thk Early History of Mankind and thk

Development of CiviLiZATtoN.
London: Murray. 1865. 8vo, vi + 378.
B, [233. i, 6]. B.M. [2378. d. 13].

Second edition : London. 1870. Se§ (26). Third edition, revised : London.
1878. (94). American edition I New York. 1878. (95). German translation ;
Leipdg. 1866. (7).

376

A BIBLIOGRAPHY OF

1865. (6) On the Negro- European DialeotB of Surinam and CoraQao. British
Association, BiriDiDgham, 1865 ; Section of Geography and

Ethnology.
Brit. Ass. R., p. 130 in section.
(delation of grammar to vocabalaij ; valae of philology as aid to ethnologj*)

1600. (7) Forschungen Uber die Urgeachichte der Uenschheit nnd die
Eatwickelnng der Civilisation. Au9 dem Englischea von
H. Miillen

Leipzig: Abel N.D. [1866]. 8vo. iii4-490*

B.M. [10006, de. 17],
*'(8) On the Origin of Language.

Fortn. Rev., 4: 544-559 (April, 1866),

(Interjectiooal and onomatopoeic words; facial exprenion and Tocal sounda;

colouring of words : sjmboliam in language— vocalization of demonetratiTea, Ac. :

names for father, mother: change of meaning; * primaeval tongaea.')

-^9) The Beligion of Savages.

Fortn. Rev., 6 : 71-86 (August, 1866).

(* The theory which endows the phenomena of nature with personal life might

be called Animism,*

Animism ; breath, phantom, dreams ; sacrifice of animate and inanimate, to

goda and man^^; origin of sacrifice— Mythology; gender; personificatloa—

Animism; guardian spirits; departmental deities; supreme Being ordering

hierarchy; Animism and science; fetishism; animism = elementary religion;

spiritualism and astrology.)

(10) Review of Memorials of Service in India, from the correspoDdence

of the late Major Samuel Charters MacpherBon, C.B. ; London,

1865.
Fortn. Rev,, 4: 639-640 (April, 1866).
(Khond gods ; human sacrifice.)

(11) The Science of Langnage : Review of Lectures on the Science of

Language ; Max Muller ; 4th edition ; London, 1864 ; and other

works.* (Anon.)
Quart, Rev., Ul) : 394-4S5 (April, 1866).
*A Dictionafj ofEngli9h Etymology ; Wedgwood : London, 1859-1 S05. Ckapien
on Language; Farrar: London, 1365.

(Development of school of Max Mnller ; his main tenets discussed and compared
with those of other writers— Evolution of Science of Language; compariaon
of words and of grammatical forms ; study of Sanskrit ; its relaUon to other
Aryan languages; study of grammar; regeneration and decay; Humboldt*s
work— Descent of languages; borrowing of words; value of comparison and
▼oeabularies — Classification of languages; Humboldt*s scheme; root words-
Origin of Language ; onomatopoeic words ; interjections - lAUguage as eridenoe
of race and of culture.}

(12) Revieiv of The Lake Dwellings of Switzerland and other parta of

Europe : Keller. Translated by T. E. Lee : London, 1866.

Fortn. Rev,, 6: 765-768 (November. 1866).

(Lake dwellings — condition of inhabitants ; culture and race; religion.)

EDWARD BURNETT TYLOR

377

1866* (13) Lecture : The Arabian Kights* Entertaiamants. Wellington
Literary Institute, April 9, 1866.
Wellington Weekly News, April 12, 1866,
(14) On Phenomena of the Higher CiTili2ation traceable to a
Hndimental Origin among Savage Tribes, British Aasocia-
tion, Nottingham J 1866 ; Section of Geography and Ethno-
logy.
Brit. Ass» R.t p. 97 in section. Anthropological Rev,, 4; 394

{abstract}.
(Mythology ; Bacnfice ; future life ; fasting.)
1867. (15) Lecture : On Traces of the Early Memtal Condition of Man. Royal
Institution, March 15, 1867.
Proo. Royal InaL, 5 : 83,

(Art of counting— Nature and habits of apiiitual beinga; theory of ghosts;
funeml gifts and sacrifices— Asaociation of ideas— Practical importance of
study of lower races.)

1868- (16) Eemew of Max Miiller 3 Chips from a German Workshop : London,
1867.
Fortn. Rev.» N. S., 3 : 225-228.
(Comparative mythology.)

(17) Beoent Spanish Bomanoes s Review of The Sea Gull (La Gaviota) ;
from the Spanish of Feman Caballero ; Hon. Augusta Bethell ;
Loudon, 1867 ; and other works.

North British Rev., 48: 129-132.

(18) William von Humboldt : Review of Schlesier's Ennnerungen an
Wilhelm von Humboldt ; Stuttgart, 1854 ; and other works.*
(Anon.)

Qoart, Rev., 124: 504-524 (April, 1868).

• Wifhelm ron Httmboldt, LehenhiM und Chamkteruftik ; Haym; Berlia, 1856.
Vber die Kawi-Sprache auf der Imel Java ; von Humboldt ; Berlin, 18S6-9,
Wilhtlm ron Humboldt's Qemmmdte Werke ; Berlin, 1841-2, &c.
(Basque language.)

(19) Lake Dwellings : Remew of Keller's The Lake Dwellings of
Switzerland ; London, 1866: Le Hons L' Homme Fossile en
Europe ; Brussels, 1867 : Lubbock's Prehistoric Times ; London,
1865 ; and other works.* {Ation.}

Quart. Rev., 125 : 418-440 (October, 1868),

* Tlie Gcoiogiail Evidences of tht Antiquity of Man ; Lyell ; 3rd ed., London,
1863. LAjke HalntatiotPi and Prthisforic Bemai fts in the Turbaries and MaH
Beds of Nofihem and Central Italy ; Gaetaldi ; trans, and ed» Chambers ;
London, 1865. Habitations lMcitstrt^\ Troyon; Lausanne, 18S0.
(Herodotus; European lake-dwellers; Robenhausen; method of building;
object of pile-dwellings ; evolution of culture ; stone, bronze, and iron ages ;
invasions ; antiquity of man.)

(20) Rev^iew of A Memoir of Baron Bunsen ; Frances Baroness Bunsen :
London, 1868.

Fortn. Rev., N.S., 3 : 715-718.

378

A BIBLIOGKAPHY OF

'^ises. (21) The Condition of Prehiatoric Raoes, as inferred from observation
of Modern Tribes » Interaational Congress of Prehistoric
Archaeology, Norwich, August 21, 1868,
Trans. Internat. Congr. Prehistoric Archaeology: London, 1869.

11-25.
(Value of particular culture-features as teata of geneml level of culture ; use
of metala; pottery; textile fabrics — twisting, spiuning, weaving; weapong ;
art; building; religion— An iinism ; prehistoric races not much below modern
layages.)

*" (22) HemarkB on Ijaoguage and Mythology as Departments of
Biolog;ieal Science. British Association, Norwich, 1868 ; Section
of Anatomy and Physiology.

Brit. Ass. R., p. 120 in Notices and Abstracts.

(Evolution of culture ; numeration; vocalization of demonstratives; growth of

myths.)
1809- (23) Review of Lnhhock's Prehistoric Times: 2nd edition, Ixjndon, 1869.

Nature, 1 : 103.

(Stoae implements; eiiigle or multiple origin of civilization; religion of

savages.)

(24) Lecture: On the Survival of Savage Thought in Modern

Civilisation. Royal Institution. April 23, 1869. i

Proc. Royal Inst,, 5 : 522,

(Study of survivals; superstition -fire-drill— mediaeval witchcraft— modem
spiritualism ^savage ammism —spiritualistic phenomena; games of chance;
divination ; disease-spirits; religious ceremonies— lustration— East and West —
Western region of the dead- sun- worship— orientation ; historical and com-
parative study ol" civilization.)

(25) Lei'tiire: Spiritualistic Fhilosophy of the Lower Baces of

Mankind. University College, London, May 8, 1869.
- 1870. (26) Reseakches into the Early History of Mankind and the
T!>e\;elopment of Civilization.
2nd edition* London : Murray. 1870. 8vo. vi + 386.
(27) Primitive Man. Loois Figuier. (L'Homiue Primiiif; Paris, 1870.)
Revised translation. Edited by E. B. T.
London : Chapman & Hall. 1870, 8vo. xix-h348.
B [221.6.74]. B.M. [2248. e.].
.^ (28) The Fhilosophy of Religion among the Lower Baces of Mankind.
Ethnological Society, April 26, 1870.
Joura. Ethn. Soc, N. S., 2 : 369-379 (abdroAt), Nature, 2:18.
(Doctrine of spiritual beings defined as Antmimt : starting-point of philosophy
of religion of lower races and minimum definition of reUgion— Soul combines
ideas of ghost and vital principle ; cause of phenomena of life, disease* dreamsi
possession ; extended to animals and inanimate objects - Phenomena of nature
explained on animistic theory; polytheism ; greater deities ; duiUism ; Supreme
Deity.)

For development of thiory o/Ammiim gee {% (15), (24) and FrimUive CuUurt (30),
wriaeninlHrn. See htter{Sb).

EDWARD BURNETT TYLOR

379.

1870, (29) Taunton College School.

Nature, 2 : 48-49.

1871. (50) Primitive Culture: Researches into the Development op

Mythology, Philosophy, Religion, Art, and Custom.
2 vols. Londoo: Mmxay. 1871. 8vo. x -f- 453 and viii + 426,
Preface, March, 187L
B. [223. i. 77. 8]. B.M, [10007. cc. 11].

Second edition: London, 1873, See (41)* Third edition, revised: London,
189L (189). Fourth edition, revised: London, 1903. (252). i^et American
edition: New York, 1874. |58). Russian translation : St Petersburg, 1872.
(83). German translation : Leipzigi 1873. (42). French tranelation : Paris^
1876. (69). Polish translation : Warsaw, 1896, 1898. {220).

(31) Article Ethnology, by the late J. C\ Prichard, Esq., M,D. ReviBcd
by E, B. Tylor,

In A Manual of Scientific Enquiry, prepared for the use of Officers
in Her Majesty's Navy and Tiuvellers in General. Published
by Authority of the Lords Commissioners of the Admiralty :
4th edition ; London, 187L 8vo. Art. IV, p. 233.

See (161).

(32) Maine'B "Villag© ComznunitieB : Menieiv of Maine's Village
Communities in the East and West ; London, 1871 ; and other
works.*

Quai't. Rev,, 131 : 176-189 (July, 1871).

* On the AffHcultumf Community of the Middle A^€S\ Nasse ; tram. Oavry:
London, 1871. Le.^ Onvriers Emvpiensi Le Play: Pane, 1855,
(Maine's poaition on theory of primitive society; patriarchal family; dis-
integration of village community system under pressure of war, increase of
population, improve' me nts in agriculture,)

1873, (33) Pervobytnaya Kultura: izsledovaniya razvitya mythologii, philo-
sophil, reHgii, iskusstva i obyiSayev. Translation edited by
D. A. Koropdevski.
2 vols. St Petersburg: Office of the Znanie. 1872. 8vo.

XV + 383 and 484 -I- viii + i^ ii.
Editor's preface, and preface of let English edition tmnslated.

(34) Etlmology and Spiritnaliam : Reply to A, R. Wallace's review of
Primitive Culture in Academy, February la, 1872.

Nature, 5 : 343.

(Were-wolves ; hypnotic influence.)

(35) Qnetelet's Contributiona to the Science of Man. Revieuf of
Quetelet*s Physique Sociale ; Brussels, 1869; and Anthropo-
m^trie; Brussels, 1870.

Nature, 5: 358-363 (see 11 : 129), Pop. So, Mo., 1 ; 45-55. North

British Rev., 22 : 217.
(Regularity of human actions; relation between biological and ethical pheno-
mena ; method ; laws of variation.)

380

A BIBLIOGRAPHY OF

1872. (36) Eastlaa's Ethnological CoUections ; Revieiv of Baatiao's Die

Rochtsverhaltnisse ; Berlin, 1872 : and Beitrage zur Ethnologie ;

Berlin, 1871: Ethnologische Foi'schungen ; Jena, 1871.

Acad. 3 : 353-4.

(Avoidance.)

(37) Lecture : Musical Tone. Wellington Literary Institute,

February 23, 1872.
Wellington Weekly News, February 29, 1872.
— — — — same lecture, Taunton College School, March 14, 1872-
Nature, 6 : 6-7.

(38) Six Lectures on The Development of Belief and Custom among

the Lower Raeea of Mankind. Royal Institution, April-
June, 1872.

(39) Lecture: The aurTival of Savage Thought in Modem Life.

Literary and Philosophical Association, Bath, December 6,

1872.
Bath Pamphlets, in Shum*s Collection of Bath Books (forthcoming),

vol. 40: Art. 11.
Bath Chronicle, December 12, 1872. p. 3,
(* Survivals V; evolution of arta— musical bow, fir&drill, Need^fire — social
position of the aged among savagea— killing of aged in Brazil, Sweden, J
England, Silesia, Saxony; Village Community— game-laws— land tenure.'
Survivals in religious observanceB— dancing— funeral ceremonies.)

(40) Report of the Arctic Committee of the Anthropological Institute :

II. Enquiries as to Religion, Mythology and Sociol<^y of

Esquimaux Tribes.
Joum. Anth. Inst., 2 : 297,
1873. (41) Peimitive Cultlke : Researches ikto the Developicekt of

Mythology, Philosophy, Religion, Language, Abt,,^ and

Custom,
2nd edition. 2 vols. London : Murray, 1873. 8vo, viii + 502 and

viii + 'izo.
Prefaces 1871 and 1873.

(42) Die Anfange der Cultur; Untersuchungen liber die Entwicklung

der Mythologie, Philosophie, Religion, Kunst und Sitte. Unter
Mitwirkung des Verfassers ins Deutsche libertragen von J. W,
SpeDgel und Fr. Poske.

2 vols. Leipzig : Winter. 1873. 8vo. xii + 495 and iv + 472.

Preface 1871 translated ; tranBlator's preface to German edition, September^

1872.

(43) Primitive Society : arts. 1 and 2.

Contemporary Review, 21: 701-718 (April, 1873); 22: 53-72
(June), Eclectic Magazine, N. S., 17: 641-652; 19: 722-740,

Tranalated, Znanie (St. Petersburg), 1873.
(1, Moral condition of savages ; treatment of nged : force of custom ; marriagi?-
prohibitiona ; savage morab under favourable conditioni ; origin of moral
standard— religion and future life— * independent morality*; kinihip, th«

EDWARD BURNETT TYLOR

381

family, blood-covenant^ oomraon mealj homicide , head-bun ting; theft; utili-
tarianiBm.

2. Blood-revenge in Australia, South America, Europe; suLatitation of
law ; lex talionis in Abyssinia ; vengeance on criminars family ; family
responsibility; evolution of idea of le^l punishment. Primitive land-law;
agriculture —digging-stick ; Village Community, disintegration under modem
conditionB. EfFect« of war on Bociety— patriarchal government 8U|)er8eded by
elective chiefahip ; slavery. Evolution of Society*)
Review of Clodd'a The Childhood of the World : London, 1873.
Nature, 8 : 99-100.
Revieiv of Herbert Spencer*s Descriptive Sociology, No. I ; English :

Collier; London, 1873.
Nature, 8 : 544-547,
(Merits of Mr. Spencer*§ scheme.)
Reniew of Marahaire A PhrenologiBt among the Todas : London^

1873,
Nature, 9: 99-lOL

(Morality ; Toda marriage ; phrenology ; infanticide*)
Englisli Bictiooaries ; Review of A Dictionary of the English

Language; Latham: London, 1866-70; and other worka.

(Amon.)
Quart. Rov., 185 : 445-481 (October, 1873).

letter on same subject Athenaeum, Nov. 22, 1873*

Lecture: The Primitive Social ConditioQ of Man. Leeds Philo-

Bophical and Literary Society, February 4, 1873.
Reports Leeds Phil, and Lit. Soc., 1872-3 (title only).
On the Halation of Morality to ReMgion in the Early Stages of

Civilization. British Association^ Bradford, 1873; Department

of Anthropology.
Brit Ass. R., 148-150. Nature, 8: 498-499.

(Growth of ethical religions; marriage; wife-purchase, &c. ; disease-flpirita ;
exordfim ; manes-worship ; future life ; great deities.)
Hefraction of Iiight Mechanically Illuatrated : questions arising

out of a lecture at Taunton College School.
Nature, 9 : 158*159.
Report of the Committee on the preparation of brief forms of

instructions for travellers, ethnologists, &c. British Associa-
tion, Bradford, 1873.
Brit, Ass. R.J 482-488.

Belfast, 1874.

Brit Ass, R., 214-218.
1874. (53) Pbimitive Culture: Researches into the Development of

Mythology, Philosophy, Religion, Language, Art, and

Custom. ^

iBt American, from the 2nd English edition, 2 vols. New York:

Henry Holt and Company. 1874, 8vo. viii + 502 and viii + 470.
Prefaces 1871 and 1873.

1873, (44)
(45)

(46)

(47)

(48)
(49)

(50)

X873
1874

(51)
[(52)

382

A BIBLIOGRAPHY OF

1874. (54) Ttie Stone Age, past and present : in Half-hoar Recreations in

Popular Science, Series I. D.C.Eates. 1874, &c. 8vo. pp- 327-862,
B.M. [8708. bbb, 4],

(Stone Age defined, p. 327 ; use of natural rtonee, 328. Unground Stone Age,
329; Drift or Qoaternary deposits; geological time: characteristic dr
implementB ; compariBon with modem and undated implemeDts. Bone Ca?a
332; Bhell-heapB; atate of ciTriUzation represented.

Ground Stone Age, 834; characteriatica ^ celt«» hammers, opears; «tOD
implements as evidence of intercourse— polished celts of green jade
Australia, 337 ; mere in New Zealand » aimilar weapon in Pern, S38 ; world-
wide uniformity of certain types.

Races without knowledge of metala ; Stone Age hi America ; copper, gold,
silver, bronze; legends of foreign teacliera; Asia, 343. Stone Age in A«ia,j
344-349 : in Europe, 349 ; linguistic evidence, 350 ; Semitic evidence, cer
monial uses, circumcision, embalming, blood-covenant. 351-355, Africa: us#
of iron, 355 ; Stone Age, 356, 357 ; thunder-bolt myth, 357-361 ; Concluflion.)
(65) Sectioiis in Notes and Queries on Anthropology, for the use of

Travellers and Residents in uncivilized lands* Drawn up by

a Committee appointed by the British Association for the

Advancement of Science.
London: Stanford. 1874.
xviii. History, 27-28, zx. Etymology, 36-37. xiii. Arithmetic, 38-39,
ixviii* Morals, 47-48. xiix. Covenants, Oaths, Ordeals, 49. xxx- Religion,
Fetishes, &c., 50-58. xrxi. Superstitions, 59. xuii. Magic and Witchcraft,
6D-61. xxxiii. Mythology, 62-63, xxxvi. Customs, 66-67. xxxvii Taboo,
67-69. Ixix. Language, 114-116. Ixx. Poetry, 117. Ixxi. Writing, 1 17-118.
Ixivii. Fire, 125. xci. Conservatism, 137. xcii. Variation, 137-138. xciii.
Invention, 138.
See (197), (236).

(56) The Relation of Baoe to Species.
Nature, 11 : 129. lUusti-ation.
See (35).

(Diagrammatic representation of effects of variation ; crossing of EuropeaniJ
and negroes ; development of a species.)

(57) The Philology of Slang.
Macmilkn'a Magazine, 29: 503-513 (April, 1874)* Eclectic Mag.,

N. S.. 19: 722-732,
(Slang a genuine and influential branch of speech; principled of formatioaj
exemplified— onomatopoeic words— abbreviation and contraction— agglutina-^
tion— variation of internal vowel*— metaphor, puns— antiquarian slang-
foreign sources.)

(58) Review of Ftiimh'E Die Eingeborenen Siid-Afrikas : Breslau, 1872.
Nature, 9 : 479-482.
(Racial types ; language.)

(59) Review of Peschers Volkerkunde ; Leipzig, 1874.
Acad., 5: 665.
(American culture borrowed.)

(60) Revieiv of Herbert Spencer's Descriptive Sociology : No. 2. Ancient

Mexicans, Central Americans, Chibchas, and Ancient Peruvians ^

EDWARD BURNETT TYLOR

383

Scheppig. No. 3. Types of Lowest Races, Negritto Races, and
Malayo-Polyneaian Races: Duncan. London^ 1874.

Acad., 6: 298.

(American and Asiatic calendars,)
1874, (61) Review of Catalogue of the Anthropological Collection lent by
Colonel Lane Fox for Exhibition in the Betbnal Green Branch of
the South Kensington Museum : Lane Fox ; London, 1874,

Acad,, 6 : 460.

(Shields; hoomemngB; spear-tb rowers ; bow; blow-gun; single or multiple

origin of inventions.)

(62) The Deganeraoy of Man : Letter on Peschel's Volkorkunde, p. 137.
Nature, 10: 146-147, 205.

(Yon Martius on degeneracy of Brazilian Indians ; BraziUan land-law ; Poly-
nesian numerala as evidence of degeneracy,)

(63) Lecture: Light. WeHiEgton Literary Institute. March 3, 1874.
Wellington Weekly News, March 5, 1874.

1875t (64) Article Anthropology.

Encyclopaedia Britannica, ninth edition, 2: 107-123.

(Man's Place in Nature— Origin of Man — Races of Mankind— Antiquity of

Man — Language— Development of Civilization ; survival in culture.)

(65) Review of Bancroft*8 Native Races of the Pacific States, vol. i :
London, 1875,

Acad., 7 : 428-429.
(Eskimo; Asiatic migrations.)

(66) Renieiff t>/Gerland*s Anthi^opologische Beitrage: Halle, 1875.
Acad., 8 r 555-556.

(Food and the origin of civilization ; fixity of race types.)

(67) Letter on Orientation.
Times, July 15, 1875, p. 7.
(Survivals in Cbrietian ritual. )

(68) Lecture : Frimitive CiviliEation. Leeds Philoiophical and Literary

Society.
1876. (69) La Civilisation Primitive: traduit de Fanglais sur la deuxl^me

Edition par M"^* Pauline Brunet.
. 2 vols. Paris: Reinwald et Cie. 1876. 8vo. xvi + 584 and

viii+a97.
Prefaces 1871 and 1873,
-^(70) ^rtic/e Cannibalism.

Encyclopaedia Britannica, ninth edition, 4 : 807-809.
(Etymology; natural aversion to cannibalism; causes of cannibalism — famine
—fury or bravado— morbid aftection— magic— religion— habit; prehistoric
evidence.)
^^(71) Article Bemonology.

Encyclopaedia Britannica, ninth editiotit 7 : 60-64.
(Etymology— Demon modified human soul; animistic theory of disease;
pOBsession ; phantoms ; guardian spirits ; sorcerers, witches, familiar spirits ;
spiritualistic stances— Hierarchy of spirits; dualism; degradation of deities
of hostile religions.)

A BZSLDTjf^LkBST OP

jbia. die ^widL Paeiie : Laadgm^ ZXH; mm

£i I

jyilt : Jail %&&, ^amiluB* iSSA. 2^ W^u

TT) £ei!rv»r 4f^ WHaca i Pmukock ICia: msv ^nsaoK; TiTrinn, 1876L

Xtfe&M: 1878.
J^wrtL A nth Inoc^ 6 : »-4f:.

/» <<iK««tM«: — KAV? 4f kcra KC actrift— &zzt Sjod;

(77) BerU^ €f TOO Hellwmld's CnhBrgcfldudtse : Aiq^dnng. ISTS.

fTtfxati6m of k-xmaa nind : cosrade boc (i-Mfftfii vidi patnliBBftl deweat.)
<7:^; lU^u^r. fjf BftDcrofc'a The Xadve Bae» cf die Fbdfie Slates,
Tola. ii-T : London, 187S-6.

Acad., 10: 192-IW.

/M^zicaa ftad Omtral Amerieaa calendar!^ mei> fte.; ]iioaad4iaflden ; Ana

jukd M«xi<y>; coafemon of fias; 5one mjthologj ia Ameriea; MezieaB

iMloge ^Coscox).;
(7fi; lUflri^j: </ Hartmann's Die Xigrider, part I: Beriin, 1876.

Aca/L, 10: 621>-€30.

^F«]lahf M reprefentatiTtt of ancieat Egrptiaai ; relatioa of Beiben ; origia

of Egjpiiaat ; Africaa migratioa iato Eorope.)
(00) Lecture : Ordeala and Oatha. Royal Institotion, April 7, 1876.

Proc Royal Inst., 8 : 152-166. Times, April 10, 1876.
{H\) ■ Macmillan's Mag., 34: 1-11. Pop. ScL Ma, 9: 307.

Living Age, 130 : 220. Eclectic Mag., N. S., 24: 59.

(Ong^n of ordeals and oaths— magic and religion. Ordeai — magical basis;

rcligioas element imported ; salt, shell-fish, water^rembling, weighing, Bible-

and -key, water, fire, poison, hot iron, food. Oatk^ nature of; oaths by son,

head, weapon; symbolic reversal; animal wonhip; 'oath of conditional

favour*; nature of penalties; mundane and post-mundane oathB, lower and

EDWARD BURNETT TYLOR

385

Mgber ciiltiire ; compound nature of Engliali judicial oath— * halidome * — *80
help me God ' ; administration to children ; social value of oath ; truth.)

-^1877. (82) Article Eimuoh.

Encyclopaedia Britannica, ninth edition, 8 : 667.
(Oriental ; ancient ; modern ; aecetic.)

(83) Sound Vibrations of Soap-Film Membranes,
Nature, 16: 12.

(84) Mr. Spencer's * PrinoipleB of Sociology * ; Review of Herbert
Spencer's The Principles of Sociologyj vol. i ; London, 1876.

Mind, 2 : 141-156 (April 1877).

(Mr. Spencer*a treatment of the evolution of religion, and Tylor*i theory of
Primitive Animism ; mental condition of primitive man ; fetishism ; belief in
soul; future life, land of souls, Hades, the West, Heaven; belief in spirits;
demon modified soul ; disease ; sacrifice ; animal-worsbip, names of ancestors ;
poljrtheistic gods ; euhemerism ; hjpotheais of verhal tnisunderstandings.)

(85) Letter, May 2, 1877.

Acad., 1} : 392.

Answer to Mr. 8pencer*s letter of April 23 ; Acad,, 11 : 367,

(Animism ; date of publication of Fhilosophtf of Edition (see 28) and writing

of Primitm Culturt*}

(S6) ■ Letter, May 19, 1877,

Acad., 11: 462 (May 26).

Answer to Mr. Spencer's letters of April 28 and May 7 ; Acad., 11 : B67, 416.
(Theory of Animism brought before Ethnological Societiy In its complete
elaboration, 1870. See 28.)

(87) LeUer, May 28, 1877.

Mind, 2: 419-423 (July).

Answer to Mr. Spencer's letter of April 19 ; Mind, 2 : 415-419.

(Relation of Spencer's views to Tylor's ; origin of Animism.)

(88) Letters, June 13, 19.

Mind, 2 : 429.

Answer to Mr. Spencer's letter of June 2,

(89) Bevieiv ofSquiers Peru : London, 1877.
Nature, 16: 191, 192.
(Stone circles ; round towers ; tradition! of the Incas.)

(90) The Cagots and Gypeies of France and Spain: Review of de
Rocbas' Les Pariaa de France et d'Espagne ; Paris, 1876.

Acad., 11 : 392-393.

(Cagots not heretics bat lepers.)

(91) *The Child', hj Ploaa : Revieiv of Ploss* Das Kind in Brauch und
Sitte der Volker ; StuttgaH, 1876,

Acad., 12 : 473-4, 495-6. Pop. Sci Mo., SuppL, 9 : 240-243 (1878).
(Survivals in treatment of infants; purification of women ; sanitary origin of
ceremoniea ; skull-deformation ; infanticide ; couvade» sympathetic magic.)

(92) Two Lectures : The Philosophy of Iiangaage. London Institution,
January 22, 29, 1877.

Times, January 23 (p. 6), 30 (p. 6), 1877.
M C C

386

A BIBLIOGRAPHY OP

1877. (93) Lecture : Iiamguage in its Lower Forms. Someiaetehire

logical and Natural History Society, March 22, 1877.
Wellington Weekly News, March 29, 1877,

1878. (94) ReSEABCHBS INTO THE EaRLY HisTORT OF HXKSIND AJTO THK

Develophent of Civilizatiox.
3rd edition, revised. London: Murray, 1878, 8vo. iv + 388.
BM, [2024. b.],

(95) New York : Henry Holt & Company. 1878.

" — (96) Backgammon among the Asteos.

Macmilians Mag., 89; 142-150 (December, 1878). Pop. ScL

Mo., 14: 49L
An expansion of (97).

(Gombination of dratigfata and dice: tustoxy of backgiunmon-groap in Old
World — Roman, Greek, Persian, Arabian tdb, Indian a^nayo, pachUi. America ;
paioiti ; Amtic infltience on Mexican culture— calendar, bronze, junk theory.
North Ameiica ; Sonora patah ; Iroquois Game of the Bowl, plum-stones or peach
stones^ deer-button 8 ; Algonquin pugasaing ; Huron jtu tU plat. Companion
of Nortb American and Asiatic culture.)

"""^(97} On the Game of Fatolli in Ancient Mexico and its probably
Asiatic Origin. Aothropological Institute, April 9, 1878.
Joum* Antk Inat., 8: 116-iai.

See (96) and (106).

(Backgammon— Arabic forms— pacAm; early accounts of pato!U\ Huron Game

of the Bowl — Iroquois deer-button and peach-stone game; junk theoiy of

Asiatic influt;nce.}

^--(96) A Folk Tale and Varions Superstitions of the Hidataa-Indiana.
Cooimumcated by E* B. Tylor. Folklore Record, 1 : 136-144.
(99) Lecture: The Beginnings of Exact Knowledge, Bath Literary
and Philosophical Association, November 15, 1878.
Bath Pamphlets, in Shum's Collection of Bath Books {forihcaming)^

vol. 40 : art. 27. Bath Chronicle, November 21, 1878 ; p, 7.
(Numeration and arithmetic ; measures ; calendar ; geometry.)

1879. (100) Article Giants.

EncyGlopaedia Britanmca, ninth edition, 10 : 571-572.
(Giant-legends; races of giants; theory of modern degeneracy; Btatistict;
giaut-myths accounted for.)

(101) Beoent American Antbropology : Remew of Bastian's Die Cultur*

lander dee alten Araerika ; Berlin, 1878 : Contributions to
North American Ethnology, vol i; Washington, 1878:
Matthews* Ethnology and Philology of the Hidatsa Indians;!
Washington, 1878.
Acad,, 15 I 12-13.

(102) Review of The Native Tribes of South Australia : Adelaide, 1878.
Acad,, 15: 88-9,

EDWARD BURNETT TYLOR

387

1S79* (103) Lecture : A Fortnight's Tour in Brittany. Wellington Literarj'
and Scientific Institute, January 3, 1879,
Wellington Weekly News, January 9» 1879.

(Stone monumenta ; journey of soul ; All Souls' Day ; wrestUiig^ ; tranBinission
of news ; marriage-broker,)

(104) Bamarks on the 0eographioal Distribution of Oames. Anthro-
pological Institute, March 11, 1879.
Joum. Anth. Inat., 9 ; 23-29.

(Draughts; kite-ijing; cat'a-cradle ; Afliatic migrations; Asiatic and Poly-
nesian mythologies.)
(105) Lecture; The History of Games. Royal Institution, March 14,
1879.
Proc. Royal Inst., 9 : 123,

(lOG) Fortn, Eev., N. S., 25 : 735-747. Pop. Sci. Mo., 15 : 225.

Eclectic Mag., N. S, 30 : 21-30.

See (96), (97j, (103).

(Principles for determining migration or re-invention of games ; kites ; catV
crftdle— Antiquity of games -Hand-games; hot cockles; morra; odd and
even— Ball play; ball dance; hostile ball -games; stick and ball; evolution
of hockey^ Sedentary games; games of chance ; divination (PHmitwe Culture^
ch* iii) ; backgammon group ; tdb^ pachm, patoUt\ North-American gambling-
gamea ; cheaa-g roup— Irregular course of evolution.)

(107) Address to the Bepartmeat of Anthropologyi British AssooiatioOt

Sheffield, August 22, 1879.
Brit. Ass. R,, 381-389.
Joum. Anth. Inst., 9 ; 235-246. Natuie,20: 413-417. Pop. ScL Ma,

16 : 145-157 (Beeent Anthropology).

(Evidence of man's antiquity derived from race, language, and culture
independent of geological evidence ; Quaternary Man ; * primaeval * or
* primitive' aa apphed to civilization; ancient Egjrptiana^ relation to
Chaldaeans ; priority of bronze over iron ; iron in Egypt ; Metal Age ;
comparative study of law and custom ; marriage, inheritance ; comparative
mythology; myth as explanation of fact.)
.^(108) Reinarka on Aiistralian Marriage Iiaws. Anthropological Insti*
tute, December 9, 1879.
Joum. Anth. Inst., 9 : 354.
(Criticism of Morgan ; letten from L. Fison and J. Forrest.)

1880. (109) Letter of condolence on the death of Broca.

Bulletin de la Soci^te d'Anthi-opologie de Paris, 3 : 498-499.

(UO) Anniversary Address. Anthropological Institute, January 27,
1880.
Joum. Anth. Inst, 9: 443-458. Nature. 21 : 381.
(Anthropology a generation ago— Review of year's work ; advance in ftomato-
logyi philology, archaeology ; recent works ; papers read before the Institute ;
marriage by capture ; Vedic religion ; fetichism.)

(Ill) On the Origin of the Plough and Wheel-Carriage. Anthropological
Institute, February 24, 1880.
C C 2

388

A BIBLIOGRAPHY OP

1880. Joorn* Anth. Inst, 10 : 74-82 ; plates and woodcuts. Revue d'An-

thropologie^ll: 718. Nature, 21: 459-460. Pop* ScL Mo., 18 : 448.
lOrigin of agriculture ; digging^stick ; hoe ; plough derived from hoe ;
development — Wheel-carriage derived from rollere ; evolution of wheel ;
war^chariot ; use of cattle.)

(112) * additional note on war-chariot.

Jouni, Anth. Inst, 10 : 128,

(113) - letter in answer to suggestion made by Mies Peacock*
Acad.. 18 : 347,

(114) Lecture I Musical Tone. Wellington Literary Institute, April

16, 1880.
Wellington Weekly News, April 22, 1880.
— 1081. (116) Antjbopolooy : An Intboduction to the Sttjby of M^ and

CmUZATION.^
London : Macmillan & Co. 1881. cr. 8vo, xv + 448,
Pi«facei February 1881 ; &elect bibliography ; 78 illuatratioM.
B. [189, f. 207]. B,M. [2352. k 23],

• New York : Appleton & Co. 1&8L 12mo- xv -f 448.

Beprinted with correctioni, 1889. See (175)* Reprinted with further correc-
tions, 1892 (196). Hepriuted 1895. RuBsian tranBlation : St. Petersburg,
1882 (127). German translation: Brunswick, 1883 (135). Spanish tram-
lation: Madrid, 1887 (163). Polish translation: Ist edition, Warsaw, 1889
(176j. Polish translation : 2nd edition, Warsaw, 1902 (245).

(116) Tlie Baoes of Mankind. Abridged from chapter iii of Antheo-

FOLOGt.

Pop. Sci. Mo,, 19: 289-311.

(117) Review o/Fieon and Howitt's KamilaiN^i and Kurnai ; Melbourne,

1880.
Acad., 19: 264-^266.
(Morgan's theorias ; origin of eiogamj ; bull-roarer.)

(118) Review of Donnan^s Origin of Primitive Stiperstitions : Phila-

delphia, 188L
Acad., 20 : 339.

(Dorman's animistic explanation of superstitions ; doctrine of future punish-
ment ; sorcery ; cannibalism ; couvade ; totemism.)

(119) Rei^iew a/ Bock's Head-htintere of Borneo : London, 1881,
Acad., 20 : 376-377,

(Value of illustrations— Day a^ a ; migration from Asia to Archipelago, Polj-
nesia^ New Zealand ; foreign influences on Bayak culture, Brahmanic, Islamic,
European ; houses, sepulchres, wood-earring, tatuing— Head-hunting.)

(120) Review of Yhear'B The Aryan Village in India and Ceylon : London,

1880.
Nature, 23 : $25-526.

(Common fields ; patriarchal family ; hereditary professions ; gnmtfl of lands ;
feudalism.)

(121) Review of Dawson's Australian Aborigines: Melbourne and

London, 1881.
Naturt!, 24 : 629--630.
(Marriage-regulations; avoidance; niimeralB; ttoiy of Pleiades.)

EDWAKD BURNETT TYLOR

389

lasi, (122) Review of Bastian's Die beilige Saga der Polynesier: Leipzig,
1881.
Nature, 25 : 28-29.
(Poljnesian civilization ; Maori coamogODy ; moon-mjtti«)

(123) Letter : Frimitive TraditdoEie as to the Pleiadea.
Nature, 25 : 150-151.
(A reply to Mr. Justice Haliburton'a letter. Nature 25 : 100, lOL)

(124) Lecture i Problems in the Hietory of Cirllizatioii* London Instil
tution, January 24, 188L

Times, January 26, 1881, p. 12.

(125) A&aiversary Address. Anthropological Instittifce, January 25,
1881.

Journ, A nth* List., 10 : 440-458,

(Review of recent work ; sign language ; burial cuBtoms ; use of atone imple-
inents ; bypcrtrichoais ; tailed men ; relation of Australian and Bravidian
languages ; survival in Scotland ; communal marriage and avoidance, refor-
mation theory ; papers read before the InHtitute ; Pitt-Rivers Collection,)

(126) notes on the Asiatic Belationa of Folyndsian Culture. Anthro-
pological Institute, November 22, 1881,

Journ. Anth. Inst, 11 : 401^04. Revue d'Anthropologie, 12: 563,
(Dyak architecture ; nose-flute ; Asiatic and Polynesian mythology ; swan
maiden.)

1882. (127) Anthropologiya ; vvedenie k iaySeniio delovka i teivilizatsii*
Translated by E. C. Evena.
S. Petei-sburg. 1 882. 8vo. xxv + 434 4- ii.

Editor's preface to Russian edition ; select bibliography ; 78 illiiatrations,
^128) The Study of Customfl.

Macmillan'a Mag., 46: 73-86 (May, 1882).

(PMloaophical speculation applied to customs— Mr. Herbert Spencer's Cere-
m&nial In^itutions ; Japanese custom of wearing two swords ; shaking hands ;
tatuing ; hair-cutting of criminals ; mourning colourg ; Methods of dealing
with evidence :— historical method— horae at funeral ; geographical method —
nose-flute ; inferential method— days of the week, sabbath, seven planets.)

(129) Itevieiv o/ Elton's Origins of English History: London, 1882.
Nature, 25 : 501-502.
(Hesperides ; races of England ; customs of inheritance ; survivals ; Mithra.)

(130) Revimv of 8hway Yoe's The Burman : London, 1882,
Nature, 26 : 593-595.

(Importance of knowledge of native habits; anlmiatic view of dreams;
propitiation of spiiits ; Buddhist doctrine of Karma and transmigration ;
Buddhist morala; Hindu influence on Burma; dancing as espression of
emotion; tattoo.)

(131) letter on the * tattooed man \

Nature, 27: 6.

(132) Lecture: A Visit to Athens. Wellington literary Institute,
January 13, 1882.

Wellington Weekly News, January 19, 1882.

890

A BIBLIOGRAPHY OF

1882* (13d) F(mr Lectures on The History of CoBtoms and Belielk. Royal
Institution, April-May, 1882.
Times, May 18, 1882. See (127).

I. April 8. The Study of CaBtoms ; its diflScultiea, and the entrust woTthisets
of speculative explanations —CriticiBm of Mr, Herbert Speiic€r*e mode of
treatment— A vni] able meibodB of tracinf^ the origin of customB— Hiitorical
method; example, the funeral horse-Bacrifiee^Geogmphical method; example^
the nose- flute— Inferential method ; example, the week and week-days.

II. April 25. Travelling of ideas and customs from Old World centres of
civilization into the Far East, Polynesia, and America— Hindu marriage-
ceremonies— Use of cycles in magic and time-meaauring— Early astronomy;
the Seven Heavens— The Elements and Ages of the World— Means of distin-
guishing between diffusion and re-invention of thoughts and arts.

III. May 2. Origin and migration of myths— Metaphor and mythical fancy —
Key to nature-myths in phrases, riddles^ &c. — The Raising of the Land— The
Symplegades— The World swallowed up by Night— The four cardinal points —
Deluge traditions.

IV. May 9. Means of distinguishing ideas and institutions in early stages of
growth, from broken-down remains of higher culture— The Polynesian and
South African probl erne— Early conceptions of nature— Animism— Effects
of early social, philosophioal, and religious ideas in the civilized world—
ConcluBion.

(134) Lecture: Original and Borrowed Oirilisatloii* Leeds Philoeo-
phical and Literary Society, November 28, 1882. Beportfi Leeds
Phil, and Lit Soc 1882-1883 (ti«e only).
1883* (135) Einleitong in daa Studium der Anthropologie und Civilisation*
Deutsche autoriBirte Auegabe von G. Siebert . • •
Bnmflwiok : Vieweg. 1883. 8vo. xix+538.
Preface 1881 ; preface to German edition ; 78 illustrations.
B.M. [10007. g. 34].

(136) Article Magic.
Encyclopaedia Britannica, ninth edition^ 15 : 199-206*

(Origin of term; magical beliefs and practices— Magic of lower races;
magicians ; fetiches - Egyptian, Babylonian, Assyrian magic ; divination ;
astrology — Greek and Roman magic ; omens — Philosophic and thenrgic
magic ; names and spells— Asiatic magic, Hindu, Tibetan, Chinese — Ma^e
in Christendom— Origin of anagic ; association of ideas.)

(137) Article Mexico (ancient).
Encyclopaedia Britaimica, ninth edition, 16 : 206-214.
(Name— History ; migration from Asia and Polynesia ; Asiatic inSuence on
civilization ; calendar, aodiac, four ages, planetary spheres, patoHi. {Se^ 97.) —
Picture-writing; native chronicles— Civilization; government; palaces; war —
Religion : mythology ; ieocalU ; prayer ; incense ; fasting ; festivals ; mock-
ing — Picture-writing ; calendar ; sacred fire— Education ; marriages ; funerals
—Agriculture and food ; clothing and ornaments ; metal-work ; art and pa»'
time — Central American culture; architecture,)

(138) Tivo Lectures on Antliropology. University Mufieum, Oxford^

February 15, 21, 1883.
Oxford Mag., 1: 88, 111. (February 21, 98, 1883.) Nature. 28:
8-11, 55^69. Science, 1 : 625 j 2 : 57.

EDWARD BURNETT TYLOR

391

I. ETolution and anthropology ; barrow -builders of England ; races of Fiji,
cranioiogj^ gramnmr ; hair and race ; pigmentation— Tjpe and environment;
development of culture ; ages of stone and metaJ ; survivals ; dreams, offerings
to dead— Geography of negro and negrito ; types of white race ; inJuenoe of
Egypt.

IL Borrowing of culture ; Eskimo ; degeneration of culture ; Veddas; evolu-
tion of culture ; sign language— Magic ; divining-rod ; worm knot ; astrology —
Anglo-Israelites— Claims of anthropology ; the Pitt-Rivers Collection.

1883. (139) Old Soandinavian Oiyillsatioii among the Modent Esquimaux.

Anthropological InBtitute, June 12, 1883,
Joum. Anth. Inst., 18: 348-357; Plates* Revue d*Anthropologie,

13 : 534.
(Early accounts of Esquimaux ; xriii. century authors ; costume ; blubber
lamp ; nith songs ; games.)

1884. (140) Artide Oath.

Encyclopaedia Britannica, ninth edition, 17 : 698-702,
(Origin of tenn ; nature of oath ; history ; classification of forma ; swearing
in early Christendom ; profane swearing ; survivals of heathenism ; political,
ecclefiiastical^ and legal oaths in modem civilized nations.)

(141) AHide Ordeal.

Encyclopaedia Britannica, ninth edition, 17 : 818-820.

(Etymology — Processefl of divination; appeals to corpse; ordeal combined

with oath ; cursed food and drink ; fire, water, hot iron ; combat.)

(142) Jntrmluction to Samoa a Hundred Years Ago: George Turner;

London^ 1884* 8vo.
B. [2068. e, S]. B.M. [2374. b. 16].

{Rapid changes ; development from spirit to deity ; Tongo incarnate in the
owl ; ceremonial survival of cannibalism ; communism.)

(143) Memoir of George Bolleston, in Scientific Papers and Addresses

by George Rolleston, M,D., F.R.S.| arranged and edited by
William Turner, M.B., LL.D., F.RS. 2 vols. Oxford,
Clarendon Press. 1884. 8vo- Reprinted for private circula-
tion*
B. [S. Nai Sci., 395a]. B.M, [2251. e, 5].

(144) Arcliaaoiogy in Ameiica.
American Architect: Boston; 16: 151,

(145) JRevieiv of Im Thumbs Among the Indiana of Guiana : London,

1883.
Nature, 29 : 305-307.

(Clasaification of tribes ; pile-honees; animism; law of vengeance (Jn^nafivia) ;
magicians ; epiritualifitic B^ncea.)

(146) American Aepacts of Anthropology: Address to the Section of

Anthropology* British ABsociatiorij Montreal, Auguet 28, 1884.
Brit Ass. R, 899-910. Nature, 30: 448-457, Science, 4: 217

{ahstraH). Pop. Sci. Mo., 26: 152.
(Flint implements ; antiquity of man ; palaeoHthle man in Asia ; Eskimo ;
Scandinavian exploration — Origin of Americang ; migi-ation from Aeia ; land
connexion ; uniformity of type ; Asiatic and American languages ; social
system ; matriarchy, matrilineal descent, rdea of residence, eiogamy, ayoid*

392

A BIBLIOGRAPHY OP

Alice, toiemism— Aiu&tic and Mezicaii comiogoiiiet and calmdari; nagie ;
f&ioUi and paehigi—DhinhKi^on of ctdture-planta and ptMmy in Kottli and
South America ; northward drift of civilization —Anthropologif in Canada)

1884. (147) North Ameriean Baeea aad CiTiliaatioii. Section of Antliro-
pology, American Association for the Advancement of Science,
Philadelphia, Penn.

Science, 4 : 345.

(Asiatic migration into North America ; diveraitj of American laagnagei ;

antiquity of man in America ; tocial condition : maternal descent.)

(148) Gttntoma of North Americaa Tribes. October Meeting of the
National Academj of Sciences, Newport, R. L, 1884*
Science, 4 : 396.

(North American and Asiatic castoms; tradition or re-invention, 'homogenj *
or * homoplaay '.)

^* (149) How the ProblemB of American Anthropology present them-
aelTes to the Engliah mind. Anthropological Society of
Washington, October 11, 1884.
Science, 4 : 545. Post, Washington, October 12, 1884. Nation&I

Bepublicaa, Washington, October 13, 1884.
(Conterratism in America ; Society of Friendfl ; Memnonited— North Amentum
ceremonies ; Ute and ZufSi boll-roarers ; re-invention or historical connexion ?
Iroqaoifl and Zoni picture-writing ; the heart line ; Mojave bark girdle ; Ute
funeral custom ; water, »oul-bridge»— Development in civilization ; Pitt-Rivers
Museum; Chri^tj collection ; systematic collection and arrangement — Boreaa
of Ethnology — Position and practical value of anthropology; * tnryivals * ;
effects of political bias,)

(150) Lectures as Reader in Anthropology in the University of Oxford.
University Museum, 1884.
Hilary Term : Development of Civilization^ — Arts of Life.

(Flint implements and their us^, with practical illustrations. Oxford Mag.
2: 20.)

Easter Term : Development of Arta and Sciences.
(Gesture langiiage — gesture and voice— in tezjectional and onomatopoeic lan-
guage — the study of language in its relation to Anthropology — art of
counting. Oxford Mag., 2 : 228, 247, 265. 298.)
Michaelmas Term : Intelleotual Development of ManMnd.
(The Pueblo Indians and their culture— Ethnology of North America— writing.
Oxford Mag., 2 : 371. 394, 410, 452.)

1885* (151) Archaeology : Appendix to Anthropology ; Daniel Wilson ; New
York, 1885. 8vo. (Humboldt Library, No. 71.)

(152) The Patriarchal Theory : Review of McLennan^s The Patriarchal

Theory ; London, 1885.
Academy, 28 : G7-68-

(Theory of the patriarchal family twenty years ago; development of
McLennan*8 views ; Primiiitfe Marriage ; matriarchal theory ; Maine^s
Jlieories of Primitive Sodetif, 1883 ; the Roman paternal family ; pafria
potuta9 and agnation.)

EDWARD BURNETT TYLOR

393

IS86. (153) Revieiv of Eourke's The Snake-dance of the Moquis of Arizona :
London, 1884.
Nature, 31: 429-^30.

(154) Amarioan Anthropology : Revieiv of Ten Kate's Reizan en

Onderzoekingen in Nord Amerika; Leyden, 1885; and other

worka.
Nature, 32 : 593-5f>4.

(Classification of Pueblo tribes ; social organixatioii ; antiquity of man in
America ; stone implements.)

A reply to Dr. Redhouse^a Netea (wi Prof, E, B. Tylof^s Arabian Matriarchaie,
propounded by him as Ffrmdent of the Anthtvpological Section o/ the British
ABBoeiaHoHf MonUtal, 1884, in Journal of Rojal Aslattc Society.
(Foaition of maternal uncle in Arabia.)
Dr. Redhouse replied, June 29. Acad., 28 : 14,

(156) ' httev, August 7.

Acad., 28: 105,

(Wilken'a letter in Bijdmgeti tot de Taal-, Land- en Volkenkunde fttn

Nedtrla ndsh -Indi^, )

(157) Report of the Committee on the acientific examination of the
country in the vicinity of Mount Roraima in Guiana. British
Association, Aberdeen, 1885.

Brit Ass. R., 690.
1885- (158) Reports of the Committee . . . appointed to investigate the
1888. physical characters, languages^ and industrial and social condi-

tions of the North-Westem Tribes of the Dominion of Canada.
Appointed at Montreal, 1884,
British Association.

I. Aberdeen, 1885 : Brit. Ass. R„ 696-708. II. Birmingbam, 1886 ; B. A. R.,
285. Ill, Mancbester. 1887 : B, A. R., 173-200. IV. Batb, 1888 : B. A R.,
233-255. V. Newcastle, 1889 ; B. A. E., 797-893. VI. Leeds, 1890 : B. A. R.,
553--715. VII. Cardiff, 1891: B.A.R, 407-449. VIIL Edinbui^h, 1892:
B. A. R., 545-615, Short report, Nottingbam, 1893 : B. A R., 653. IX. Oxford,
1894 ; B. A R., 453-463. X. Ipswich, 1895 : B, A.R„ 522-592. XL Liver-
pool, 1896 : B. A. R., 569-591. XIL Toronto, 1897 : B. A. R., 791 (title), and
B.A.R.. 1898, 628-688.

(159) Lectures as Reader in Anthropology. University Museum,
Osiord, 1885.
Hilary Term : I, II, HI, Early History of the Arts and

Sclenoes.
(Numeration— Weights and Measures.)

IV, V, VI. Passages in Herodotus relating to Anthropology.
(The Lake-dwelleR — the Scythians— funeral rites.)
Oxford Mag,, 3: 44, 61, 83, 104, 126, 149.

Easter Term : Development of Mythology^ Magic, Games, &o.
Michaelmas Term : Social and Beligious Systems.

394

A BIBLIOGRAPHY OF

(Degrees of rel&iionBbip — origin of religiooi ideas in primiiiTe aocielx^

Animism— Animism and mythology.)
Oxford Mag-, 3: 387, 405, 443.

1886. (160) Article SftlutatioM.

Encyclopaedia Britannica, mnth edition, 21 : 235-237,
(Embrace ; kiss ; earesses ; crouching, prostration, kneeling, bowing ; un-
covering ; grasping hands ; words of greeting.)

(161) Article Antiiropolo^ in A Manual of Scientific Enquiry:

5th editioo ; London, 1886,
Art, vii, p. 225.
See (31).

(162) Lectures as Reader in Anthropology. University Mudeom,

Oxford, 1886.
Hilary Term : Mankind— their diatributioa, antiqiiity, and early

condition.
Easter Term : Origins of Civilisation.
Michaelmas Term : Modes of Exprefision^GeBture-&ig&% natural

sounds, pictures; language, writing.

1887. (lS3) Antropologia : Introducci6n al Estudio del Hombre y de la

Civilizaciiin. Trad ucida del ingl^ por Don Antonio Machaday

Alvarez . . .
Madrid: Falcdn. 1887. 8vo. xiii + 529.

Preface, 1881 ; special preface to Spanish edition ; select bibliognipliy ;
77 illustrationt.

(164) Review of Lang's Myth, Ritual, and Religion : 2 vols. ; London,

1887.
Acad., 32 : 277.
(Origin of myth in early stages of human knowledge.)

(165) Account of a * Witches' Iiadder ' found in Somi^et. British

Association, Manchester, September 2, 1887 ; Section H.
Brit. Ass. R, 900 {iUle only)* See Folklore Journal, 5 : 1-5,

(166) Lectures as Reader in Anthropology. University Museonit

Oxford, 1887.
Hilary and Easter Terms : Development of Arts as illustrated in

the Pitt-Bivers Museum.
Michaelmas Term : I, II, III. Development of Missile Weapons ;

IV, V, VL Origins of Ornamental Form and Decoration.
1887- (167) Reports of the Committee appointed to edit a new edition of
1892, Anthropological Notes and Queries. British Association.

I. Manchester, 1887 : Brit. Asa. R, 172. II. Newcastle, 1889: B.A.R, 186,
III. LeedB, 1890 : B, A.R., 547. TV. Caniiif, 1891 : B. A. E,, 404. V, Edin-
burgh, 1892: B. A. a, 537.

1868. (168) Jniroduction to Aino Folk Tales: B. H. Chamberlain ; London^
1888. 8vo. viii+57.
B,M. [Ac, 90S8/11].
(Earl J accounts of Aino ; phjBieal characteiigtica ; Alno langoaget place-name!

EDWARD BURNETT TYLOR

396

in Japan ; mythology, Japanese and n&tive elementfl ; folk-lore and primiti?e
philoBopby.)

(169) I7et66 on Powhatan^s Mantle, preserred in the Aslnnolean

Muflemn, Oxford.
Internationales Archiv fiir Ethnograpbie, 1 : 215-217. Plate.

Nature, S9 ; 232.
{Powhatan used as Algonquin tribe name and title ; Tradescant^s Virginia
collection ; all ell-work.)

(170) Sayagee : Letter on The Arlthmetio of the Lower Baoee.
Times, September 17, 1888.

(Decimal syatem founded on use of digits.)
^^171) On a Method of Inveatigating the Beyelopment of Institutions
applied to Itaws of Marriage and Descent Public Lecture,
Oxford, June 7, 1888.
Oxford Mag,, 6 : 43L Joum. Antk Inst, 18 : 91-92.
(172) —^ - ■ British Association, Bath; Section H. September 7j

(173)

1888.
Brit. Ass. R, 84a TimeB» September 8, 1888.

' Anthropological Institute. November 13, 1888.

Joum, Antb. Inst., 18 : 245-2r2, Nature, 30 : 143.

(Avoidance, relation to cnstomsof residence, teknonymy ; levitate ; convade —
Priority of maternal system ; transition from maternal to paternal system ;
marriage by capture— Relations of exogamy and claasificatory Byetem ; croes-
cousin marriage ; relation of capture and exogamy ; origin of exogamy in
political con&iderationa.
In diacu^ian .*^ statistical principles.)

(174) Lectures as Reader in Anthropology. University Museum,

Oxford, 1888.
Hilary Term : Anthropological Elucidation of Passages in Greek

axid X»atin Authors.
Easter Term : Eaoes and Languages of the World.

Public Lecture : June 7, See (170).

Michaelmas Term : Baoe» Language, and CiviEzation*

— — Report of the Reader in Anthropology for 1888.

.^Oxford Univ. Gaz., 19: 405.
1889. (175) Anthrot OLOQY r AN Introduction to the Study of Man and

Civilization.
2nd edition, revised. London: Macmillan & Co.; and New
York, 1889. cr. 8va xv-l-448.
Preface, 1681 ; select bibliography : 78 illnatrations.
(176) Antropologia. Wst^p do Badania Cziowieka i Cywilizacyi

Translated by Aleksaodra B^kowska.
Warsaw: Naklad tygodnika * Prawda '. 1889. 8vo. xi +
413 + ii.
Preface, 1881; select bibliography : 78 illn strati on s.

396

A BIBLIOGRAPHY OF

1889. (177) Letter : The Fertilization of the Date-Palm in Ancient Assyria ;

May 31, 1889.
Acad., 35 : 396.

(178) Kotes on the Modern Survival of Ancieat Amulets against the

Evil Eye, Anthropological Institute, March 12, 1889.
Journ, Anth. Inst., 19 ; 54.
(Origin of face^braaeefl in Latin phalnixe.)

(179) Lectures as Reader in Anthropology. University Museum,

Oxford, 1889.
Hilary Term i The Anthropology of the Higher 19'ations.
Eaater Term: The Anthropology of the Higher Nations —

Aryans.
Michaelmas Term : Development of Eeligions.

Report for 1889. Oxford Univ. Gaz., 20; 396.

1889- (180) Ten LeHure^ on IS'atural Beligion. Gifford Lectureship : Uni-

1890. versity of Aberdeen, December 1889 and January 1890,
December, 1889.

L letrodiictor J— Antiquity and Early Condition of Man. Natural Re-
ligion, treated by methods of Natural Science, forma a branch of Anthropology
or the Science of Man, so that an introductory aecoimt of this science becomes j
neceasaiy— Anthropology restg on the principle of the blgh antiquity of Man.
The division of the human species into Racea, the formation of Families of
Languages, and the nae of Civilizations, being shown by the monumenU of
Egypt and Babylon to have already taken place as early as 3000 to 5000 B.C.,
a long period antecedent to this is required— Means of estimating this period
given by Geology. Time required for rivers to excavate their valleys, since
the early ages when Man inhabited Europe together with the Mammoth and
other extinct mammals— Condition of Man in the Palaeolithic period illus-
trated by his rudely-chipped stone implementa, contrasted with the finer and
polished implements of the more modem Neolithic period— Problem how far
the state of savages of the early Stone Age is represented by that of savages
survivinij into modern times— Stone implements of Tasmania compared with
those of European drift- men— Analogy between modem and remotely ancient
savaged as hearing on the Development of Civilization.

IL Development and Tranamisston of Culture. Through all branches of
Culture, development depends on similar operations of mind. It ia thus I
possible to learn, from invention and progress in m^iterial arts, the principles
to be applied to the study of intellectual and religious institutions— Inventions
do not arise by sudden and spontaneous efforts, but by gradual and progressive
modiScations— Illustration from the history of fire-arms— The directing ten*
dency in development is often that of practical utility^ but it may be some
other motive^IUuatration from head-dresses of Friesland— Development
cannot be safely conjectured from a single stage, but requires study of the
series of stages- Illustration from the history of ball-games— Method of
determining whether similar arts or customs in different regions were inde-
pendently invented, or derived from a common source— Illustration from
certain games allied to backgammon in the Old and New World,
111. Natural Theology and Natural Beligion. The term * Natural Theo-
logy ' due to Varro, the Roman grammarian— Eaymund us de SsbundA- Expe- ,
diency as an avowed ground for belief in past ages, now discountemukced-^ ^

EDWAED BURNETT TYLOK

397

Theory of the invention of religion for political purposes tm tenable — ^Riee of
the theory of Natural Religion as implanted in, or reasoned out by, Primitive
Man ; Herbert, Vobs, Wilkin s, Clarke, Bntler— This Bo-called Natural Eeligion,
defined aa mainly consisting of Monotheism and the doctrine of Future Retribu-
tion, is an artificial extract from the religions of cultured nations, and in no
way corresponds with the actual religions of the lower raceB— Scheme of the
philosophic framework of the religions of the world : its b&aii in Animism or
the Doctrine of Souls—The religions of low and early races exemplified in
that of the Tasmanians ; comparison of its doctrines with those familiar to
the modem civiUied world.

IT. Souls. The Human Soul as defined in early stages of culture — ^The
Australian race aa representatives of PrehiBtorie Man— The Soul conceived
as breath, blood, shadow, &c. ; these terms not originally metaphorical, but
expressions of primitive science — ^The Life-soul and the Phantom-soul ; their
combination ; the doctrine of Several Souls of Man—The Life-sonl ; its depar-
ture the cause of death, trance, sickness, sleep, and dreams— The Phantom-
soul ; its appearance in dreams and visions— Development of the early
doctrine of Soul, and its representation in Art.

¥. Soule {continued). The Phantom-soul ; its appearance in dreams and
visions— Souls of AnimalSi Plants, and Objects generally— Materiality and
Mortality of the Soul, as defined in Early Animism— Egyptian religion ;
earliest recorded conceptions of Immateriality and Immortality— Later
definitions of the Soul in Classic and Christian philosophy— Rise of indepen-
dent Psychology.
January, 1B90.

VI. Futni^ liife. Existence of the Soul after death— Ohoats haunting or
returning to their home or burial-place ; efforts of the living to propitiate
or expel them ; funeral sacrifices and feasts— Departure of the soul to a
distant country, mountain, or island ; Western location of Region of Souls j
Underworld or Hades ; Heaven — Life of Departed Souls— Doctrine of Con-
tinuance — New application and survival of Funeral Sacrifices— Rise of the
Doctrine of Retribution ; its history,

¥IL Spiritual Belngt. Spirits pervading the world— Demons ; regarded
as souls of the dead, or as of similar nature— Their functions as causes of
actual events ; their division into Good and Evil —Demons as causes of disease ;
the doctrine of demoniacal Possession and Obsession constitutes a primitive
Theory of Medicine— Cure by esorciem— Inspiration by Demon ; Oracle-
possession— Familiar spirits — Belief in Witchcraft involves early explana-
tions of real phenomena— Its supersesgion by Science.

VI I L Spiritual Beings {continued). Souls of ancestors regarded as Patron-
Spirits^ Honse-Spirit— Guardian Genius— Festival of the Natal Genius ; its
survival in the modem world— Winged Spirits— The Winged Figures of
Assyria — Fertilization of the Date-Pal m— Hebrew Cherubim— Greek and
Roman genii— Angels in Christian Art.
Nature 41 : 283. Science 15 : 12ft.

IX. Nature Spirits. Nature regarded in early religion aa animated by
Souls, or Spirits similar to Souls— Nature -Spirits ; Watei^Spirits, Nymphs,
Nixes ; Tree and Forest Spirits, Elves, Fairies— Nature Deities ; Heaven,
Earthy Sea, Sun, Moon, Rivers, Mountains, Ac— Polytheistic Deities; their
anthropomorphic character ; their transmission from one national religion to
another— Supremacy in Polytheism,

X. Folytheism- Paiittieisin— Monotheiam. Doctrine of the Supreme
Deity in Polytheism ; its distinction from Monotheism— Introduction of

398

A BIBLIOGRAPHY OF

belief in a Supreme Being anaong uncultured races from ChriBiianitj and
Mohammedanism ; the * Great Spirit ' — Good and Evil Principles in Dualism —
Rise of Pantheism and Monotheism— Connexion of Religion and Philosophy,
See (188) 1890-91.
ISOO. (IBI) Preface to Ling Roth's The Aborigines of Tasmania: Halifax*
1890.
B, [2471&4. d. 3]. B.M, [10492. f, 82].

(Tasmanians as representatives of palaeolithic man; stone-implement-making ;
comparison with man of Mammoth Period in Europe.)

(182) SuppUmeidary Note : 2nd edition ; Halifax, 1899.

B.M. [10492. ff. 24].

(Progress made in anthropological study of Tagmaniana ; implements, colleo-
tton in Pitt-Rivers Museum, Oxford ; comparison with European implements —
acrapers— plateau flints of Kent ; traces of palaeolithic civilization in Wett
Australia and New Zealand, Set Eariy Histojy of Mankimlf p. 195.)

(183) Revievj of White's Ancient History of the Maori : Iiondon, 1889.
English Historical Review, 5: 391.

(Maori myth, tradition^ genealogy ; Maui myth and matrilineal pedigree.)

(184) The Winged Figures of the Assyrian and other Ancient

Montiments. Society of Biblical Archaeology, June 3, 1890.
Proc. Soc. Bibl. Arch., 12: 383-393. Plates.

(Assyrian and Egyptian winged figures ; * tree of life ' ; fertilisiation of date*
palm ; significance in Assyrian nature-worship : honeysuckle pattern ; influ-
ence on Hebrew^ Classical » and Christian art.)
See QiflFord Lecture, viii. (180).

(185) Lecture: Anthropology. Bristol University College, Septem-

ber 30, 1890.
Western Daily Press, Evening News, Bristol Times, Bristol
Mercury, October 1, 1890.

(Relation of anthropology to mechanics — stone implements, origin of
handles; decorative art— palm and honeysuckle pattern; physics— magnet ;
Animism ; Tasmanian religion.)

(186) Lectures aa Reader in Anthropology. University Museum,

Oxford, 1890.
Hilary Term : Development of Beiigions.
Easter Term : Barly Bevelopment of Institutions : Family and

Tribe ; Property ; Penal Law ,- Government.
Michaelmas Term ; Elementary Anthropology*

Report for 1890. Oxford Univ. Gaz., 21 : 50L

1800- (187) Report of the Committee appointed to investigate the habitSt
1802. customs, physical characteristics, and religions of the native^

of India.
British Association j Leeds, 1890.
Brit, Ass. R., 547.

Edinburgh, 1892*

Brit Ass. R., 615-617.

EDWARD BURNETT TYLOR

399

1800- (188) Ten Lectures on natural Beligion. OifFord Lectureship: XJni-
1801. versity of Aberdeen, December 1890 and January 1891,

December, 1890.

I. Spirit Interoourae. Communication of Spiritaal Beings with Man in
Dreami and Visions; Incubation -Oracular PosBeaaion and Obsesaion ; In-
Bpimtion— EemainB of the Dead inhabited by the Soul; relic-worship—
Images associated with, or inhabited by» Spirits : Idolatry — Transition from
the Animistic doctrine of Spirit acting on Matter to the scientific doctrine of
Force.

II. Bitea and CereinoziieB. Image-worship as a means of communication
with deities— Prayer and Sacrifice as exemplified in the religions of lower
and higher races.

III. Bites and Ceremonies {candud0d). Fasting— Narcotics— Temples and
Shrines— Feativals^Rites kept up in sport.

IV. Magic. Magic based on Association of Ideas— Symbolic practices;
bewitching by images, &c. Sympathetic Medicine ; Evil Eye ; Divination--
Astrology ; Days of the Week — Relation of Magic to Animism and its place
in Development of Thought.

V. Ideas of the Universe. Barbaric and ancient theories of the Universe ;
Heaven ; Hades— The Sun's descent at night into Hades ; its disappearance
behind mountains— Upraising of Sky ; Su negates— Planetary spheres ;
Buddhist, Moslem, Christian systems,

January, 189L

VI . PrimitiTe Society. Marriage-systems ; maternal and paternal family ;
Marriage by capture, service, purchase ; Exogamy— Introduction of religious
control into marriage institutions.

VI L Primitive Sooitty {contintied). Primitive Morality independent of
Religion; its apparent irregularity due to difference in circumstances and
stages of development— Kinsfolk and strangers; infants; the aged and
incurable — Direct and indirect acts — burying, setting adrift, immuring.
VI I L Primitive Society (conclucUd). Transformation of Social Conduct —
Gift and Trade ; Vengeance and Criminal Law — Collective ReBponiibility^
Individualism— Introduction of religious control over moral and social laws.

IX. JMffUsion of Bites and Beliefs. Rosaries ; Jain, Buddhist p Hindu,
Mohammedan^ Christian— Modern diffusion of accounts of a Deluge-
Geological myths — Methods of distinguishing native from imported ideas
in the religions of the world.

X. Causation in the Universe* Transition from Spiritual to Physical
Theory of Nature ; Newton's Principia— Theological and Philosophical
Causation; Free-will and Necessity in the earlier and later stages of
culture— Conclusion.

See (180).

188L (189) Primitive Culture: Researches into the Development of
Mythology, Philosophy, Relioion, Lanouaqe^ Abt, and
Custom.

3rd edition, re viged. 2 vols. London (Oxford pr.) ; Murray. 189 L
8vo. xii + 502 and viii -f 471.

Prefaces, 1871 ; 1873 ; September, 1891. Paging of 2nd edition retained.

B, [247115. d. 38, 39]. B.M. [101)07. cc. 25].

400

A BIBLIOGKAPHY OF

ISdl. (190) History of Marriage : .Review o/ Wefitermarck'fi History of Httman
Marriage ; London, 1891.
Acad., 40: 288-289.

(Use of tenn * marriage ' ; relation of Wefltermarck*! views to ^hom oiVmrwim
and A. R. Wallace ; * promiscuity * ; Tjlofw own poeidon on ^irMhmm of
primitiTe society i»e€ I7S); statistical method; prohibitton of mama^
between kindred; exogumj; aversion; patriarchal and maternal syitetts:
sexual selection.)

-..^(191) The Limits of Savage Boligion. British Association, Cardiff;
Section of Anthropology. August 21, 1891.
Brit Afls. R.» 800. Nature, 44 : 511.

(192) ^ Anthropological Institute, November 10, 1891.

Joum., Anth. Inst, 21 : 283-301. Nature, 45 : 71. Times, Novem-
ber 21, 189L
(Errors of obsermtion ; Great Spirit ; Mandan delnge-myth ; gods in South
America; ideas of fntare life— West Austalian Motogon and Chenga ;
Baiame and other deities of South-east Australia; ideas of fntare life—
Tasmanian belieffi*)

(193) Charma and Amuleta. International Folk Lore Congreas, 1891.
Trans, Intemat Folk Lore Congress : 387-393.

(GuUwin from Australia; onion stnck with pins; €orp-cn\ witches* ladder;

worm'knot; cimarutQy &c*)

(194) Discussion on General Pitt-Rivers' paper on Typological Museuma.

Society of Arts, December 16, 1891.
Journal of the Society of Arts, 40 : 121.
(The Pitt-Rivers Mnsenm ; use of typological collections.)

(195) Lectures as Header in Anthropology. University Museum,

Oxford, 1891.
Hilary and Easter Terms : Elementary Anthropology.
Micbaelmas Term: Origin and Development of Language and

Writtng.

Report for 1891. Oxford Univ, Gaz., 22 : 394.

I8d2. (196) Anthropology: a.n Intboduotiox to the Study of Man and

Cty iLizATr oii^
3rd edition revised. London: Macmlllan & Co. 1892, cr. 8vo,
(197) Sectiona in Notes and Queries on Anthropology : edited for

the Council of the Anthropological Institute by John George

Garson, M.D., and Charles Hercules Read, F.SA.
2nd edition^ London : the Anthropological Institute. 1892. 12mo*

X + 242. Coloured Plate.
First part entirely re-ca«t; second part rented and additional chapten
written.

xTi. Fire, 112. itvii. Indention, 113. imii. Variation, 114. xx. Conserra-
tism, 116. xxii. Writing. 118. xxrii. Religion, Fetishes, Ac, 130. xxviii.
Mythology, 140. xxix. Superstitions, 142. xxx. Magic and Witchcraft,
144. ixxii. Morals, 146. xxxiii. Covenants, Oaths, Ordeals, 149. xxsr.

EDWARD BURNETT TYLOR

401

CiiAtoniB, 151. xlii- Etymology, 170, rliii. Language, 171, xliv. Poetry,
174, ilv, Hi8tor}% 175. Ix. Arithmetic, 20^.
Se4! (55) 1874, (236) 1899.
1802. (198) * Couvade '— the geneeis of an anthropological term: Lette7\
Nov, 2, 1892.
Acad., 42: 412.

Answer to letter from Dr. Murray. Acad., 42 : 389.
(Authority for term in French literature.)

(199) Letter, Dec. 6, 1892.

Acad., 42 : 542.

(Basque and Bearnese cou\'ade,)

(200) AnniTersary Address. Anthropological Institute, January 26,

1892.
Jonrn. Anth. Inst, 21 : 396-41 L

(Review of recent work ; plateau implements ; origin of art ; Malagaty art ;
position of Berber lang^ioges; anthropological work of Wilken and Moseley).

(201) The Tasmanians as representatives of Frehistorio Han« Ash*

molean Society, Oxford, May 30, 1892.
Oxford Mag., 10: 377 {notice).
See (207) 1893.

(202) A Stone Age Basis for Oriental Study: Inaugural AddreflS,

Section of Anthropology : Ninth International Congress of

Orientalists : London, September, 1892.
Trans. Ninth Internai Congress Orientalists, 2 : 805-813. Smith-

soDian Reports, 1893 : 701-708.
(Palaeolithic Egypt; Tasmanian implement-making; Tasmanian moiab,
language, religion, not below normal savage level ; permanence as well as
development j^jsaible in culture— Neolithic culture; South Sea UlanderB;
aborigines of BeluchiBtan and China; Hindu and Egyptian traditiouB—
Migration of culture from higher to lower nations ; Asiatic influence in
Polynesia and New Zealand.)

(203) Lectures as Reader in Anthropology. University Museum,
Oxford, 1892.

Hilary Term : Early Stages of Literature and Science.

Easter Term : Early Stages of Science and Art.

Michaelmas Term: Anthropology as related to Ancient and

Modem History.

Report for 1892. Oxford Univ. Gaz., 23 : G26*

1893. (204) Polk Rhymes: Reuievj of Baring- Go uld*s Strange Survivals;

London, 1892: and Northall's English Folk Rhymes; London,

1892.
Academy, 48 : 73*
(205) Anniversary Address. Anthropological Institute, Januaiy 24,

1893.
Journ. Anth. Inst., 22 : 377-384.

(Beview of recent work ; importance of stnicture of language ; Blackfeet
Indians, phone tics ^ kinship terms, &c«)

402

A BIBLIOGRAPHY OF

laoa. (206) Bemarks on a CoUeotion of the Rude Stone ImplementB of the
Taam&nians, showing them to belong to the Palaeolithie or
nngronnd stage of the implement^maker'a art. Aiitiiropo-
logical Institute, Mai-ch 21, 1893*

Journ. Anth. Inst, 23 : 52. Nature, 47 : 527.

(Moral level of TaamaniansO
(207) On the Taflmanians aa Bepreeentatiyee of Pa laeo lithic Xan.
Anthropological Institute, March 21, 1893.

Joum. Anth, Inst, 23 : 141-152. Platea.

See (201) 1892.

(Earlj accounts of implemei^td ; miBeialogical data ; method of

handling, u«e— Comparison with European palaeolithic implemeats; sezapen ;

bafling^ Degeneracy of Tasmamana— Summaiy of Tasmanian ctdtuze.)

(206) Lectures as Reader in Anthropology. University Hv

Oxford, 1893.
Hilary Term: Anthropology of Social and Political Inatitu*

tiona.
Easter Term: Anthropology of Moral and Heligioaa Inatitn-

tiona
Michaelmas Term : Bacee of ManJund aa classified by Languago^j

CiTiliaationi and History.

— Report for 1893, Oxford Umv. Gat, 24: 581,

(209) Report of the Committee on Unlfonnity in the Spelling of San

and Barbaric Languages and Race Names. Britisli

Nottingham, 1893*
Brit Ass. R., 662,
1884- (210) T. S. Hnxley as AnthropologlBL
Fortn- Rev., N. a, 58: 310-311.

(211) On the I>if[\ision of Mythical Belieft as Svid^noe in tho

of Cnltnre. Biitish Association, Oxford ; Section H. Aiigii8^|

9,1894.

Brit Ass. R, 774. Nature, 50 : 439.

(Aiialic iafluence on pre^olumbian cultm^ of Amerijca ; Briiige of the Deal;
Mexiesii and Buddhist JoumeT of the SouL)

(212) On some Stone Implements of Anstr^ian Type firom Tasmania^

British Asaociationi Oxibrd ; Section H. At2gnst» 1894l
Brit Ass. R., 782.

(213) On the Occurrence of Ground Stone Implements oC AnatraUaji

Type in Tasmania. Anthropological iTiiititnitr^ Deoember 11»

1894.
Jonin. Anth. Inst, 24 : 33&-M)l Plate.
(ComparisoB with Earopeaa psketilitliir fisjigawrals: ^silioa of haHisf ;
deschption of speciinens.)

(214) Ladiirea as Reader in Anthropology. UiiiTemtjjr

Oxford, 1894.

EDWARD BURNETT TYLOR

403

Hilary and Easter Terms: Baoes of Mankind as olaaamed by

Language, CiTiligation^ and History.
Micherelmas Term : Intellectual Development of mankind,

Language, WritlDg, Arithmetic, &;e.

Report for 1894. Oxford Univ. Ga^., 25 : 546.

1806. (215) Pleyte'B Batak Tales : Remeiv of Pleyte*8 Bataksche Vertellingen ;
Utrecht, 1894.
Acad., 47 : 308-309.

Times, March 30, April 6, 1895.

I, Accounts of the religions of savage and barbaric races— Effects of intercoufae
with the civilized world — Communication with foreign nations proved by
Deluge- traditions, ^c—Animigm of lower races reduced to its native
elements,

II. Classificixtion of Religions under the Animistic sjstem— TaBmaniana as
modem representatives of Prehistoric Man in the Palaeolithic period —
Tasmanian Animiam^American Indians as representatives of Prehistoric
Man in the Neolithic period— American Animism — The Classic or Graeco-
Roman Stage of Animism — Comparison with the great religions of the
modem world.

(217) Lectures aa Reader in Anthropology. University Museum, Oxford,
1895.

Hilary Term i Early Stages of Knowledge — Science, Magic,

Mythology, History.
Easter Term : Anthropology of Social and Political Institutions.

(218) Lectures as Professor of Anthropology in the University of Oxford.
University Museum, 1895.

Michaelmas Term : The Belation of Savage Life to the Higher

Forms of Civilisation.
— — ^ Report for 1895. Oxford Univ. Gaz., 26 : 579.

(219) Public Lecture: The Patriarchal and Matriarchal

Family Systems. University Museum, Oxford, October 24,
1895,

1806, (220) Cy wilizacja Pierwotna. Badania rozwoju mitologji,filozofji, wiary,
mowy, sztuki i zwyczajdw. Translated from the 3rd English
edition by Z. A. Kowerska (Madame Rzijd) ; with preface, bio-
graphical sketch, and notes by Jan Karlowicz.
2 vole. Wai-saw: The * Glos' : R Czemak. 1896, 1898. 8vo.
iv + 433 and iii + 416.

Editor*8 preface ; biography; prefeices, 1871, 1873, 1891; notes.

B,M. [10007. dd. 6],
(221) Introduction to The History of Mankind. An English translation
from the 2nd edition, 1894-5, of Ratzel's Volkerkunde; Lon-
don, 1896.

B.M. [70O2. h, 3].

(Value of illustrations; of mua«um specimens; gnjuping of ethnographic

D d 2

404

A BIBLIOGRAPHY OF

ooUections; Pitt Rivers Museum. Decorative art of saTa^es— Native and
borrowed culture ; study of material culture valuable for studj of social,
moral, and religious inatitutiona— State of anthropological science^

1B96. (222) On American IfOt-Gtames, as Evidence of Asiatic Intercourse

before the Time of Columbus.
In EtbDograpbische Beitrage « * . dem Professor Adolf Bastian

gewidmet zn seinein TO***" Qeburtstage am 26. Juni 1896.
Internationales Archiv fur Ethnographie, suppL to ix, pp. 55-€7.

Coloured plate, and figures.
B, [1902. c, J].

(Relation of pachm and paf of /i'— Backgammon ; Arabic form; Chineie;
mentions in Sanskrit Uteratare; later history— Early mentions of pai4xNi;
spread of game northwards ; Apache lot-g&me ; Huron game of the Bowl ;
Iroquois game of deer-buttons -Summary of evidence.)

(223 J The Matriarchal Family System,

Nineteenth Century, 40: 81-96 (July, 1896),

(Patriarchal theory of primitive family; McLennan^s Primitive Marriage i
*promisciiity*— Maternal descent; kinship; a vert ion— Priority of paternal
or maternal system — Maternal system asBOciated with residence in wife's
family ; Sumatra, Formosa, Kasia, Garo ; Pueblo Indians ; North America ;
Banzai, Ashanti; Queensland - Origin of maternal system: (1) social causes;
residence; inheritance, the Erbtochtermann ; (2) political; exogamy; peace,
numerical strength —Effect of paternal family instinct ; elopement, capture,
purchase — Regularity of human action; scientific method; advantage of
antbraixjlogy over other sciences in dealing with more direct evidence.)

(224) The Formation of the Family : Remeiv of Grosse's Die Formen der

Familic ond die Formen der Wirtlischaft ; Freiburg & Leipzig,

1896.
Nature, 55 : 51,

(Economic claasiication of mankind ; division of labour between the sexes ;
maternal clan and family ; exogamy not derived from aversion to marriages
of near kin.)

(225) Steinmetz on the Evolution of Punishment : Revieio of Steinmets's

Ethnologische Studion zur ersten Entwicklung der Strafe:

Leiden, 1894.
Academy, 49 : 49*

(Law of vengeance; formalization ; duel; family discipline ; punishment by
the community ; vengeance and chastisement in relation to religion ; ghost-
fear; future punishment and reward.)

(226) The Hale Series of Huron Wampum Belts : Notes and (uldenda

to Four Huron Warapom Kecords : a Study of Aboriginal
American History and Mnemonic Symbols ; by Horatio Hale.
M.A, (Harvard) . . . Anthropological Institute, December
8, 1896.
Journ. Anth, Inst., 26 : 248-254. Figui'ee.
(Date shown to be recent by European method of drilling beads.)

(227) Lectures as Professor of Anthropology. Univeraity Museum,

Oxford, 1896.

EDWAKD BURNETT TYLOR

405

1896, (228)

Hilary Term : Ancient and Barbaric Life as related to Higher

Civilization.
Easter and Micbaelmas Terms : Stmotiire and Development of

Iianguage ; ClaaBiBeation of Mankind by Language. Fioture-

writing and Phonetic-writing,

Report for 1896. Oxford Univ. Gaz,, 27: 634.

Public Lecture : North American Picture-writing, with

special reference to a series of historical Wampum Belts of

the Hurons. November 23, 1896.
1807. (229) Lecturt»s as Professor of Anthropology. University Museum,

Oxford, 1897.
Hilary Term : Early Stages of Knowledge — Soiencei Magic,

Mythology, History.
Easter Term : Anthropology of Social and Political Institutions.
Michaelmas Term : Anthropology of Social^ Moral, and Religious

Institutions.

— — — Report for 181*7. Oxford Univ, Gaz., 28 : 5fK).
189 8* (230) Remarks on the Totem Post from the Haida village of Masset,

Queen Charlotte Islands, now erected in the grounds of Fox

Warren, near Wey bridge.
Joum. Anth. Inst., 28; 133-135. Plate.
(Haida geneabgy.)
(231) Remarks on two British Columbian House-poets with totemic

carvings, in the Pitt Rivers Museum, Oxford.
Jom-n. Anth. Inst., 28: 136-137. Plate. L'Anthropologie, 10: 230.
(Houae-pofitB belonging to Haida-Tsimahian group of tribes; totem-myth;
Killer-Whale ; conventional art ; representation of embodied spirit ; Killer-
Whale myth and modern Indian belief in a good and evil deity.)
"^232) Remarks on Totemism, with especial reference to some Modern

Theories respecting it. Anthropological Institute, May 24,

1898.
Journ. Anth, Inst, 28 : 138-148. Nature, 58 : 189,
A Paper arising out of the preparation of an address to a meeting of a
philosophical society in Oxford under the pretsidency of Profe^aor Sanday, on
certain views on the anthropology of religion contained in the works of
Mr. J. G. Frazer and Dr. F, B. Jevons.

(Ty lor*B position on Totemism ; his first writingB on the subject, 1867 ; influence
of M*"Lennan—* Totemism,' existing confusion in terra g—M^^Len nan's PHmt'
live Muniage, 1865; totems and exogamy; Wmship of AninmU nnd Plants \
totemism proper in North America; Long's Qjibway iotam a confusion of
totem and manitu; M^'Lennan's extension of the term; origin of deities;
Florida, Fiji — Frnzer'^ Totemism fl8&7 ; Samoa; incarnation aninaala ; genera-
lization in primitive philosophy ; * apecies-deities '; Patagonian family-deities ;
Haida tot^m-clans and creation-legends— Exaggeration of position of totemism
in history of religion— Jevons' Introduction to the Bistoty of Rdi(tion\ hypo-
thetical ^totem-Bacraments *— Eecent history of subject; McLennan, Herbert
Spencer; Frazer's Golden Bough; sou I -trans fere nee, external soul in folk-
lore— Wilken'a view of origin of totems; Animism . , . of the Malay

406

A BIBLIOGRAPHY OF

Archipelago, 1884-5; animal-worship; transmigration of bouU— Ck)drington's
The Melanesians; souUtranaference — Baldwin Spencer's Tnhes of Central
A%^9tralia\ exogamoua system of the Artinta ; churinga, local totems —
Exogamy and totemism, existing independently and combined; Bocial action
of totem*)

1898. (233) The Survival of Palaeolithic Conditiona in Tasmania and Ana-
tralia, with especial reference to the modern use of ungronnd
stone implemeots in Western AnstraUa. British Association,

Bristol; Section of Anthropology. September 9» 1898,
Brit Asa. K, 1014. Joom. Anth. Inst, 28: 199.
Nature, 59 : 162.

( Quasi 'pahieolitliic character of Tasmanian implements ; character of Austra-
lian implements recently palaeolithic; use of chipped stones in Australia;
ground stone hatchets introduced from Torres Straits ; reaemblance between
Australian and prehistoric European skulls.)

(234) Speech on the final report of the Committee on the North- Western

Tribes of Canada, British Association, Bristol, September 10,

1898.
Nature, d9 : 162. See Brit. Aes. R, 628.
See (158) 1885.

(235) Lectures as Professor of Anthropology* University Museum,

Oxford, 1898.
Hilary Term : Ancient and Barbaric Stages of Culture compared

with Advanced Civilization.
Easter Term : Anthropology of Ancient Civilized Nations.
Michaelmas Terin: Development of Culture— Arts of Ltfe-

— Report for 1898. Oxford Univ. Gaz., 29 : 614.

1890. (236) 8ectio7W hi Notes and Queries on Anthropology ; edited for the

British Association for the Advancement of Science by John

George Garson, M.D., and Charles Hercules Read, F.S.A.
3rd edition. London : the Anthropological Institute, 1899.

12mo. xii + 252. Coloured plate.
* Edited by a Committee of the British Association of which Prof. E. B. Tyler,
F.R.S.^ is chairman.*
Sections as in 2nd edition.
See (55) 1874, (197) 1892.

(237) Lectures as Professor of Anthropology, University Museum^
Oxford, 1899.

Hilary Terra : Bevelopment of Culture — Arts of Life.

Easter Teim ; Anthropology in Classical Literature.

Michaelmas Term : Classification of Mankind by Race* Lan^uage»
and Civilization.

Report for 1899. Oxford Univ. Gaz., 30: 714.

1900. (238) The Stone Age in Tasmania as related to the History of Civiliza-
tion. British Association, Bradford ; Section H. September 5,
1900.

EDWARD BURNETT TYLOR

407

Brit Ass. R., 897. Journ. Anth* Inst., 30 : Reviews and Miscellanea,

37. Nature, 62 : G36-G37.
(Tasinaiiian culture below tlmt of Man of Mammoth Period in Europe ; stone
iiuplements, artsoflife^ mythology; animiam.)

IBOO. (239) On Stone Implemdnts ttom. Taamama : Extracts from a letter by
J. Paxton Moir. ExhibitioD of Tasmanian Implements. Anthro-
pological Institute, November 27, 190O,
Joura, Anth. Inst., 30 ; 257-262. Plates.
(240) Lectures as Professor of Anthropology. University Museum,
Oxford. 1900.
Hilary Term : Anthropology in Ancient Iiiterature.
Easter Term: Early Stages of Civilisation — Arts of Snbsifltence

and Protection.
Michaelmas Term : DevelopmeiLt of Langaage ; Writing i Arith-
metic.

Report for 1900. Oxford Univ. Gaz., 31 : 672*

1000- (241) Reports of the Committee on the Ethnological Survey of
1902. Canada. British Association.

(Nominated at Liverpool, 1896 ; reported 1897, 1898, 1899.)

IV. Bradford, 1900 : Brit. Abs, R^ 468-568. V. Glaigow, 1901 : Brit. Aia. R,

409, 410. VI. Glaagow, 1902 : Brit. Asa, K, 353-449.

(242) Enquiry as to 'Seijeant Bettesworth.' Notes and Queries, 9th
serieB, 7 ; 127.

1901. (243) Lectures as Pitifessor of Anth^opolog}^ tlaiversity Museum,
Oxford, 1901.
Hilary and Easter Terms: Social, Political, and Beligioua
Institutions of the Ancient Civilized Natioms, as compared
with those of Savage and Barbaric Peoples.
Michaelmas Term : Bace, Language, and Civilization.

— Report for 1901. Oxford Univ. Gaz., 32 : 650.

(244) Public Lecture : Totems and Totemism^ with special reference to
the Totem-post from British Columbia in the University
Museum. University Museum, Oxford, November 22.
Oxford Mag., 20 : 108.

(Origin of exogamy not in aversion but in political conaide rations,)
ike (247).
lOOa. (245) Antropologja. Wst^p do Badania Czlowieka i Cywilizacji • ••
Translated by Aleksandra B^kowska.
Second edition. Warsaw: Stefan Dem by, 1902. la 8vo. xi-i-446.
Photograph ; preface, 1881 ; select bibliography ; 78 illustrations.
(246) Anthropology: supplement under corresponding headings to
article in Ninth edition.
Encyclopaedia Britannica, tenth edition, 25 : 464-467.
(Taamanians as representativea of low palaeolithic man — stone implements.
weapOEB, food, clothing, artsi language^ numeration, religioai social life.)

408

A BIBLIOGRAPHY OF

1902. (247) Lectures as Professor of Anthropology* University Moseum,
OxJbrd, 1902,
Hilary Term : Anthropological Evidence in Ancient Literature.
Easter Term : Early Stages of Civilization — ^Language, Writing,

Arithmetic^ &c.
Michaelmas Term : Outlines of Anthropology.
' ■ Report for 1902, Oxford Univ. Gaz., 33 : 548.

(248) A Kote on a Haida Totem-post lately erected in the Pitt Rivers
Museum at Oxford.
Man, 1902j 1. Coloured plate: illustrations,
(Description ; use of word ' totemiam \)
(240) Malay Divining Bods.
Mao, 1902, 40.

(250) Memorandum on the present state and future needs of Anthro-

pology in Oxford: in Statements of the needs of the University;
Oxford, Clarendon Press, 1902. 8vo. p. 70.
B, [Oxon. 8vo. 692].

(251) Report from the Delegates of the Oxford University

Museum (signed, on behalf of the Heads of Museum Depart-
ments), Ibid., p. 84,

1003. (252) Primitive Culture: Researches into the Development of

Mythology, Philosophy, Religion, Lanouaoe, Abt, and

Custom.
Fourth edition, revised, 2 vols, London (Plymouth pr.) : Murray,

1903. 8vo. xii +502 and viii + 471.
Prefaces, 1871, 1873, 1891 ; October, 1908.

Noticeable alterationu ; I. 167— Vocal Tone ; II, 234-238— Totemiam.
B. [S. Hiflt. civ. P^J. BM, [2024. b],
(253) Lectures as Professor of Anthropology. University Museum,

Oxford, 1903.
Hilary Teim : BacOf lianguage^ and Civilisation.
Easter Temi : Development of Letters, 14'umerals, Weights and

Measures.
Michaelmas Term : Anthropological Beconstruotion of History.
Report for 1903. Oxford Univ. Gaz., 34: 560,

1004. (254) Ret^ieiv of Reports of the Camhridge Anthropological Expedition to

Torres Straits : vol v : Sociology, Magic^ and Religion of the

Western Islanders,
Manchester Guardian, October 12, 1904, p. 5, !

(Aetiological myths and folklore jMUrallela.)
(255) Lectures as Professor of Anthi^opology. University MuseutDi

Oxford, 1904.
Easter Term : Prehistoric Man,

Michaelmas Term : Anthropology in Ancient History.
Report for 1904. Oxford Univ. Gaz., 35 : 672.

EDWAKD BURNETT TYLOR

409

1904- (256) Preliminary Report of the Committee on the present state of
1000. Anthropological Teaching, British Association, Cambridge,

1904.

Brit. Ass, E.» 341 (titled

Second Report, British Association, York^ 1906.

Brit, Ass. R, 701,

(257) Committee for Anthropology : Paper 1, Memorandum on the
Position of Anthropology in the Univei-sity. Submitted to the
Hebdomadal Council, Mcbaelmas Term, 1904.

Oxford : privately printed*

(258) Committee for Anthropology : Paper 2. Memorandum on a pro-

posed Diploma Course in Anthropology. Hilary Term, 1905.
Oxford : privately printed.
IB 05. (259) Introductory note to an anonymous Obituary Notice of Professor

Adolf Bastian.
Man, 1905 : 76,
(260) Lectures as Professor of Anthropology, University Museum,

Oxford, 1905.
Hilary Term : Haoei Language, and Civilization.
Michaelmad Terra : Origin and BifTasion of lianguages,
— Report for 1905. Oxford Univ. Gaz., 36: 612.

1006. (261) Lectures as Professor of Anthropology. University Museum,

Oxford. 1906.
Hilary Term : PreMstoric Man,
Easter Term : Bace, Language » and Civilization.
Michaelmas Term : Early Stages of Art and Knowledge.
^ — — — Report for 1906. Oxford Univ. Gaz., 37 : 663.

1007. (262) Lectures as Professor of Anthropology. University Museum,

Oxford, 1907.
Hilary Term r Early Stages of Art and Knowledge.

Easter Term : Haces of Mankind.

Announced for Michaelmas Term

History (especially Classical).

Anthropological Evidence in

Secimi Anthropology in Hints to Travellers, 5th ed., pp. 222-243^

London, R.G.a, 1883,
Sett ion Anthropology in Hints to Travellers, 6th ed., pp. 371-392.

London, R.O.S., 1889.

IXDEXES

Wtgaemhthiadu^i

L SUBJECT ISBEX

U4;

ABcsdeem doiO, *au
Abigail, stejoi; 143.
AfeakyB,stej«<li^

mi, Idl, I83w
Ailoni, riftea, 192:

Mrka. milk

murder ia, Ifl^;
peoplei, lai. Sm'wim Am-
crai, Asaibtty Bafcnwi,
Bftiro^ Btato, Bajalcm

Eatkmen, GaUa, Hetiofpo-
lii, Hercso, Kabjle, Kaiv

«o, Ffnr, Soath Africa,

UkxEBi. Wakamba, Waki-

kajB, Yao.
Age giaidea, 321.
Agneahare, ongia, (111>;

xiiet, 199.
Aiao,(168X
Axr-gnn, diseharge, 18.
AMeringa, 215.
Algonquins, (169).
Altar beneath oak, 125.
Amayonii 288.

America, archaeology, (144) ;
Asiatic infinencet in, (961
(9r7> (1041 (133), (137),

(146), (14^ (211), (222);
cvH^omi, (148) ; races and
ciTilization, (147). See aleo
Algonqnins, Anahnac,
Aztec, Brazil, Eikimo,
Haida, Iroqnois, Mexico,
Peru.

American philology, (2).

Americans, origin of, 146.

AmuleU, (178), (193).

Anahnac, (1).

Anatolia and CjpTus, 21b.

Ancestor worship, 181.

Angoni, warrior's custom,
108.

Animal heroes, 185 ; -named
groups, 178 ; societies.

r,lll,(^(15).CM^
(28),(»V(7I^(»4>,i:ilS).
(1»V(M«); -
of fmrrwifciMi, 21d-£ld;
Laezntsasflii^^.
AsAzopoid soeieKies^ 346.

(161),(l^(34tf);
a4S); Amenaa
(14f): Aa IxtzodoetiMi ta
the Stadf of Maa a»i
OfOzBtioa. (115> (liT)^

fmx (i«K G7»)^ (1^

277; needs of, 377. Sw
ateOxfimL

Aatiqazty of nna, (lOfT).

i^es. faaknts ei; 345l

ApoDoaxos ok the Sgyaai,
256u

Appeflatxie gods, 81, 94.

Aiah BOBC, 237, 943, 245.

Afabiaa matriarckale (155),
(156).

Arabiaa Nights (13)l

Arehneology, (151) ; ia Amei^
ica, (144): British New
Goiaea, 325-41.

Arctic Committee, (40)l

Argos, mrntrflineal wcccesioa,
302.

Arithmetic, psimitiye, (170).

Armenian prostitiitioB, 190.

Armshells, 332.

Art, (200); decoratiTe, 332 sq.,
(185).

ArmnggmOika^ 22^.

Amnta belief, 354; custom,
146 ; primitiTenesB, 210.

Asia, eee Aino, Aiiatotis,
Arab, Assyria, Babylon,
Badaga, Burmese, Celebes,
Chinese, Chins, Cyprus,
Dyaks, Hebrews, India,
Japanese, JsTanese, Koo-
kies, Korwa, Lydian, Ma-
nipur, Medes, Mongol,
Mylitta, Persian, Tamils,
Todas, Veddahs.

Asiatic influence in Poly-
nesia, (73), (119), (126).

Association of ideas, 220, 224.

313;

195;

S5-7»;kiasy|i,

lawi,n06);

ia^JS.

AvcagerofUood,142.
AToi&Ke, 4, 345, 348» 349,

352. 353,(125). (173).
Axe, ohmdiaB, 327.
Aaimha castor, 197.
Astecs,(96X

Bahyhm, prostitntion at, 189,

194, 2oa
B^cdiQs,35^
Backgammon, (96).
Badaga caatma. 14L
Bag, Slc^ soul in, 146.
Rahima, milk costODis, 155.
Baiame, (1^>
Bairo, mUk custom, 155.
Baupoub, H., Fire-piston,

17-49.
Bantu customs, 140 sq., 197.
Bards, 293.

Barkiini customs, 208.
BartleBay, 326.
Bastian's works, (36).
Bayaka, murder among, 108.
Beast children, (3).
Bechuana custom, 161, 162.
Belief and custom, deTelop-

mentoi;(133X(188).
Bellows and flre^iston, 44.
Bendt Tah6b, 114, 120.
Betel«mortar and fire-piston,

45.

SUBJECT INDEX

411

Bible, folk-lore of tbe, 101-

174.
BinandeTe-spoaking pcoplei

miX
Bird design, 357.
Bisection of tribe, 209,
Blood and curses, S65 ; and

milk, 158; aa Hbation,

124 ; drawing, 107 ; re-
venue, 1 OS, 142,(43). (145) ;

Bhed on the ground, 104;

•wit, 107,
Blow-gun and fire-pistozif 45.
B. N. I angle, 73 ; in Anara-

lian skulls, 74 ; in S€ottiBb

skill li, 75,
Bogo castom, 1S3.
Bouing milk, 155.
Borrowing, see Monogeneait,
Brazil, murder in, lu7.
Breaking pote. 115.
Bregma angle, 67 ; in Aub*

tralian skulla, 73.
Bride lifted over threBbold,

170.
Bride-lifting, 170.
Bridge, use of, 142,
Brittany, (103).
Bronze age, spesr-headB of,

263,
Bronze and iron, (107).
Brother and sister, 52 ; avoid-

aoce, 345.
Buddhiitm, (180).
Building, stone charm, 132.
Bull, sacrifice of, 140,
Bullom custom, 154,
Bull-roarer, (117), (149),
Bundle of Life» 143.
Bu r eau, an t kropo logi cal , 28 1 .
Burial alive, 109; beneath

threshold, 172; Spartan,

305.
Burmese skulls, 'third point'

in, 70.
Bushmen, skulls, * third

point * in, 70,
Byblua, rite at, 192,

Caffots, (90).
Cain, mark of, 102-10,
Cairn cove nan ts, 131 ; cqb-
torn, 364, 368,

Calendar, (60). (78), (137).

(146).
Camel, habiU of, 349 ; milk-

ing;, 160,
Cannibftliam, (70),
Cannon and Jire piston, 45.
Capital punishment, origin

of, 105.
Carriage, origin of, (111).
Casnar, 291.

Categorical imperative, 221.
Cattle, habits of, 347,
Cattle-fold taboo to women,

159*

Cattle-kraal taboo to women,

158.
Celebes custom, 146.
Censua, sin of a, 173,
ChamiH and amulets, (193).
Chaims against evil eye,

355-9.
Chastity, sacrifice of, 189,

196,
Chief, hallows milk, 160;

house of, 168 ; taboo of,

142, 233.
Chieflains, female, 299.
Childbirth and cattle, 161 ;

and threshold, 171.
Children, origin of, Austra-

lian belief, 215; sacrifice

of, 124
Chinese custom, 168 ; music,

241 ; skulls, * third point *

in, 70.
Chins, * third point * in, 70.
Chirope, 105.
Cfco-i", 215.
CAtinw^rt, 146,211,
Cinerary urns at Cnossus, BOS,
Civilization, (14) ; history of,

(124) ; modem (24) ; ori-
ginal and borrowed (184);

primitive, (50), (69); Scan-
dinavian, (139).
Classificatoty system, 59,

30^^23.
Cleansing milk-vesaels, 157.
Club^head, 331,
CoUingwood Bay, 325 sq.
Commandments, Ten, 151.
Commensality, 373.
Communal marriage, (125).

Se€ aho Group-marriage.
Comparison of skulls, 67, 68.
Conception, ideas as to,

213 sq.
Concept ional totem ism, 217.
Conditional curses, 361-74.
Consanguine family, 311.
Consanguinity. 322,
ConsonancCf 238.
Contamination of music, 235,
Continence, 130, 162.
Conu9 shell, 330, 332.
Corinth, polygamy at, 301.
Com-god,96.
Cornucopia design. 358,
Coronation beneath oak, 127.
Counting unlucky, 174.
Cousins, mating of, 57-63.
Couvade, (77), (173), (198),

(223),
Covenants, 373; on the cairn,

13L
Cratls, (21).
Cranial development, 68 ;

vault, proportions of, 71.
Crawley, A. E., Eiogamy

and the Mating of Cousins,

51-63 ; 200, 344.

Cross- cousin marriage, 57,

Culture, (126); evolution of,
(5), (19). (21), (53), (61),
(107), (HI). (146), (202),
(207); hero, 183; plants,
(146); primitive, (30).

Cumont on religious prostitu-
tion, 201.

CuNKiNOHAM, B* J., The
Australian Forehead, 65-
79.

Curds, eating, 162.

Curses in Morocco, 361^74.

Curvature, frontal, 76 &a.

Customs, (128); and beliefs,
History. (133); of Ameri-
can Tribes, (148).

Cypriote cuetom, 189.

Cyprus, sigi^nna in, 271;
swords in, 274 ; and Hall-
fftatt, 274.

Dairymen's customs, 142, 159.
Danger, mystic, of taboo, 223.
Bate-palm, (177).
Dauko, 325 sq. ; pottery from,

335.
David and Abicail, 143.
Death as pollution, 161 ;

dances, 183; milk taboo

after, 161,
Decalogue, 152.
Decapotifl, forest of, 112.
Decay of languages, (11).
Decorative art, 3B2.
Defloration at puberty, 195,
Degeneracy, (62): of man,

(62) ; of liMmaniana. (207).
Degeneration, (100), (138).
Demonology, (71).
Demonstratives, (8).
Deimrt mental gods, 81.
Desai, 292.

Dialect, Borian, 306.
Dionysus, 98.
Disguise as protection against

ghost, 108.
Divination, (72), (81), (141),

179.
Divining Rods, (249).
Divinity, Greek conception

of, 90.
Dorians,Wlio were the ? 295-

308 ; classical, 300.
Doris. 297,

Dowry earned by prostitu-
tion, 190, 194,
Drama, ghost in, 180.
Drill weight, 335.
Drinking milk, 158 sq,
Druidical custom, 125.
Dryads. 120,
Dyaks, (119).

Eariy History of Mankind.
„ m, (7). (26). (94).
Earth spirits, 367.

412

SUBJECT INDEX

Elder and yonnger, distinc-
tion bet ween f 321.

Ellij<f HaveloGkj on incest, 52.

Encyclopaedia articles, (64),
(70), (71). (72), (82), doO),
(136), (187), 140), (Ul),
(160), (246),

Endogamy, 345 ; effect of, 57,
62.

Enemies, slain, precautions
against, 108.

*Eneti, 256 sq.

England, rares of, (129).

Engraved shells, SS2.

Eponymous heroines, 299*

Eskimo animism, 145 ;
Scandinavian civilization
among, (139).

Ethics and religion, 182,

Ethnographical muBeums,
277.

Ethnological Study of Music,
235-5i

Ethnology and spirittialism,
(34).

Euahlayi beliefs, 215*

Euhemerism* 97.

Eunuch^ (82).

Europe, «« Aberdeen, Achae-
ans, Argos, Athens. Attica,
Brittany, CagotB, Corinth,
Dorians, 'Eneti, England,
Gnesatae, Glauceeter, Goi-
dels. HaUBtatt, Hebridew,
IberianB, lllyria, Ligurians,
Neanderthal, Negro, Norse,
Romove, Scandinavian,
Sigynnae, Sparta, Thessa-
liansL

Evidence, dealing with, (128).

Evil eye, 171, (178), 355-9.

Evil power ns magic, 225.

Evolution, differential, (106).

Evolutionary arrangement of
museum, 280.

Eichange of women, 56,

Executioner's customs, 106.

Exogamy, 5, 51-63, (173),
(190), 203-10, (223), 343^54,

Expiation for bloodshed, 105.

Expulsion of ghosts, 108, 109;
young males, 351,

External soul, 145 sq.

Face brasses, (178),

Fair at Hebron, 130.

Family, 14,(223), (224); for-
mation of, (224); matri-
archal, (223). See also
Social.

Farksll, L. R., *SoniIer-
gutter' in Greek Poly-
theism, 81-100.

Fasting, protective, 106.

Fat her- daughter incest, 348.

Female monarchs, Illyrian,
299 ; warriors, 288,

Fi^orLnes, wooden, 176*

Fiji, classificatory syetem in,
59; tribal groups, 106;
warrior*6 custom, 109.

Fijian custom, 168 ; kinship,
314,

Fire-pistoD, 17-63 ; in British
North Borneo, 34 ; Bur-
mah, 23 ; Europe, 17 ;
Flores, 37 ; French Indo-
China, 27; Java. 36; Malay
Peninsula, 27 ; Pegu, 27 ;
Philippines, 37 ; Sarawak,
32 ; Sumatra, 31; — among
the Aetas, 38 ; Igorrotes, 38 ;
Kachin, 23; Kbas, 27;
Mois, 27; Sam-Sam, 29;
Sea Dyaks, 32 ; Shan, 24,
26; Wa, 24; — list of illus-
trations, 46-9; method of
using, 2H, 32; origin and
dispersal, 39; with duct,
20, 34.

FirstcouBins, marriage of, 61,

Flake, obsidian, 328.

Flattening, frontal, 65.

Flesh kept apart from milk,
164.

Fly-god, 99 ; sacrifice, 99,

Food, (66); eating, as coven-
ant, 131 ; taboos, 223.

Ford, custom at, 139,

Forehead, Australian, 65-79 ;
defined, 65,

Foster kinsliip, 317,

Frazer, J. G., Folk-lore in
the Old Testament, 101-
74 ; on taboo, 222.

Fi'ee love, 298.

FRfiiKfi-MAEREco^B., Biblio-
graphy of Tylor, 375^09.

Fresh milk, use of. 162.

Frontal angle, 66 ; bone, ele-
vation 0?, 66 ; curvature,
76 ; curve angle, 74 ; curve
index, 74, 75 ; ridge, 78.

Functional gods, 81, 94.

Funeral customs, 180; rites,
Spartan, 305.

Future life, 181, (14), (24),
(50). (84), (103), (118), (149),
(180), (188), (192), (211).

Gaesata€, 265.

Gae8um, 265, 266, 268; in

Spain, 270.
Gal la custom, 124.
Games, (96), (97), (104), (105);

American, (222),
Gawa, 327.
Genna, 228.
Geographical arrangement

of museum, 280.
Geometric ornament, 297.
Ghost, 179 ; angry, 105 ; and

ma na layer, 105 sq*
Giants, (101).

Gideon, story of, 125.
Giffbrd Lectures, (180), (188).
Gilead, forest of, 113.
Crinui, 261.
Gloucester, Nine Witches oC

285-93.
Goats, milking, 159 n«, 160,
Gods, contest with, 139.
Goidels, 289,

Goodenough Island, S25 sq.
Gorgon*B Head^ 356.
Government and anthropo-

logy, 279.

Grail e

stories, 286.

Gmphic musical records, 25S.

Glass, 141.

Great spirit, (192),

Greek beliefs, 104, 105 ; mu-
sic, 241 ; musical 8ca,l^
242; Polytheism, 81-100.

Group-brotherhood, 818 ;
-marriage defined, 316;
-marriage and kinship
terms, 315; -motherhood,
318 ; primitive human, 350.

Gwiddon, 286.

Gtcyfiit Gloyw, 290.

Haddon, a. C, Religion of
Torres Straits Islanders,
175-88,

Haida, (231),

Hair, dedication of, 192;
-dressing, 301 ; -dyeing at
Athens, 304 ; of Spartans,
303 ; of Thebans, 303.

HaUstatt, spear-heads at,
263 ; and Cyprus, 274,

Harmonic intervals in melo-
dy, 246.

Harmony, 237.

Hartlakd, E. a, Rite at the
Temple of Mylitta, 189-
202.

Hawaiian family, 311; skulls,
* third point ' in, 70.

Hebrews, sacred trees of»
110 sq.

Hebrides, oath on stone, 138,

Helio polls, prostitution at,
189.

Heraclidae, 298.

Herero belief, 110.

Hero culU, 183 ; ipreadof,186.

Herodotus and anthropolo
255 sq, ; on Dorians,
Sigynnaa of, 255-76.

Heterogamy, 848*

HeterODhony, 238.

High places, 116.

Hindu mu&ic, 245.

Holy places, 368.

Homer and the Dorians, 30^.

Homoiogamy, 343,

Horses, habits of, 349.

Hosea, tomb of, 118,

Houses of the soul, 148.

SUBJECT INDEX

il3

Howitt, A. W., 203. 3ia.
Huxley. (210).
Hybridity, (56).

lljerlanB, 270.

Illyrm, hc^me of Dorians,

295 sq,
Ilkriane, 298.

laipttrity of manslayer^ 105.
Inbreeciing. 54*
Incest at paberty, 200 ; aver-

BJon to» 344.
Incobation, 179.
India, oath on stone, 132, 183.
Initiation, 182, 320,
Inteiralfl, musical, 241.
Introductions, see Prefaces,
Invocation, 176,
Ion a, oatli on atone, 133.
Iriah invaders, 292,
Iron age, 255 sq,, 262 ; in

Cyprus, 273; Spain, 270;

problems, 275,
Iron of the Jura, 262,
Iroquois kinship, 314.

Jabbok, 136.

Jacob and Laban, 134 ; at

the Jabbok. 136 ; covenant,

131; daughiera, 114, 120;

story of, 126; wrestling,

13a
Jade axes in Aufttralia, 3.
Japanese music, 242 ; mytho-

W, (76).
Javanese custom, 132.
Jawbone as talisman, 106 n.
Jealousy, animal, 346.
Jevons, F. B., on taboo, 220.
Jewish custom, 165 ; religion,

102.
Jordan, scenery of, 112,
Joshua, story of, 126.
Joyce, T. A., Prehistoric

Objects in British New

Guinea, 825-43.
Jumping on a threshold, 167.
JusprimoM noctU^ 201,

Kabvie belief, 162.
Kakhyeen custom, 142,
Kai-ens, 54,
Kavirondo, warrior's custom,

loa

Keepers of the Threshold, 167.
Key Island custom, 146*
iC folk, 807.
Kid, seething, in its mother's

milk, 15L
King and oak, 127 ; election

of 132.
Kinship, and exogamy, 51 ; in

Torres Straits, 312 ; terms,

origin of, 320.
Kitchen midden, (4).
Knots, tying, 370, 37 L
Koi, 213,

Kookie custom, 132,
Korwa custom, 170.
Kurnai kinship, 313.
Kwoiam, 184,

Labialization, 307.
Lake-dwellings, (12), (19).
Land in Laconia, 363.
Lang, A., Austialian Pro-
blems. 203-18; E. B/Tylor,

1-15.
Language, (8), (11), Ct8), (22),

(57), (92), (93), (200), (205) ;

baby, 8,
L-'Ar, 361-74.
La Tene, 262.

Law, evolution of, (43), (225),
Lectures (unpublished), (151),

(186), (195), (203), (208),

f214), (217L (218), (219).

(227), (228). (229), (235),

(237), (240). (243), (244 .

(2471 (258), (255), (260),

(261). (262),
Leechcrafl, 370.
Legal character of exogamy,

53.
Legend, dramatized, 180.
Licence, prenuptial, 298.
Life, bundle of, 143 ; modern,

survivals in, 39,
Ligurians, 261,
Linguistics, principles, (11),

(205).
Lipped pottery, 334.
Lissauer 8 diagraph, 66, 76.
Loop handle, pottery, 339.
Lorentz's patent, 19.
Lot games, (222).
Luck, menstruoufl women

and, 160.
Lucretius on animism, 2,
Lycurgus, 301.
Lydian proatilubion, 190,
Lying-in woman and luck,

160.

Mabinogion, 285.

McLennan, (152).

Madagascar customs, 132,
142,

MaduK 176.

Magic. 199, (9), (24), (28), (71),
(81), (133), (136), (188);
and religion, 166 ; in
Torres Straits, 176; nega-
tive, 219-34 ; primitive
idea of, 225,

Maian, 185,

Maidens, auction of, 193,

Maine's Village Communities,
(32).

Malayan family, 310.

Males expelled, 351.

Man, degeneracy of, (62).

Jfona, 219 sq., 226.
Mang'anja custom, 197.
Manipur customs, 228.
Mannhardt on sacrifice of

chastity, 199.
ManslayeV, taboo of, 104*
Mansla3*er3, customs of, 103.
Maori, (103); kinship, 314;

Bkulli, * third point' in,

70.
Mark of Cain, 102-110;

Tubal, 103.
Marks, executioner's, 106.
Marett, R. R,, Is Taboo a

Ne^tive Magic? 219-34.
Marriage customs, 132, 198,

(178); of Veneti, 298;

regulationB, 4, 5, 343 ;

Australian, 203, (108).
Masai customs, 141, 160, 164 ;

milk custom, 155.
Masks, 177,

Maternal family, (173), (223).
Mating of CouBins, 57-63,
Matriarchal family system,

(223),
Matriarchate, Arabian, (155J,

(156).
Matrilineal descent at Sparta,

301.
Medes, 260.
Median dress, 259; empire,

extent of, 260; origin of

Sipynnae, 260.
Mediterranean swords, 274.
Melody, 237.
Melanesia, exogamy in, 54 n.,

58.
Melanochroi, 298.
Menstruation, 157,
Menstruous women, and

luck, IfiO ; taboos of, 187.
Metals, knowledge of, (54) ;

use of, (107),
Method, craniological, 69, 73,

76.
Methods, statistical, (173).
Mexico, (1), (97), (137).
Migrations, (119).
Milk, seething a kid in, 15L
Milking, the work of men,

159 ; the work of women,

160.
Mi Ik' vessels, cleansing, 158.
Misima, 325 sq.
Mithra, (129),

Mixing tiesh and milk, 164.
Mohammedan tombs, 116.
Mohmo, 161.
Moloch, sacrifice to, 124.
Mongol custom, 170.
Monogamy, primitive, 6,
Monogenesis, 3, 40 sq., (96),

(97). (104), (106), (126),

(133), (137), (138), 148 ,

(149), (211), (222).
Moon God, 187.

^^^H ^^r

SUBJECT INDEX

^^1

^^^^^^H Morality, (50); and a future

125 sq.; of Mamre, 127;

Pitch, absolute, 247? and ^^H

^^^^^^^H lifei 1>^2; and religion,

weeping, 127.

timbre, 248 : musical, 24 L ■

^^^^^H 182; and ntualp 152.

Oaths, (80), (140).

Placenta buned beneath ^M

^^^^^^H Morgan, L. H., 809 ; theories

Obsidian implementa, 326.

threshold. 172. ■

^^^^^H

Oceania, see Australia. Fiji,

Pleiades. (121), (123). ■

^^^^^^^H Moriori gkuUfl, ' third point*

Melanesia, Moriori, New

Plough and wheel carriage, H

^^^^^^^1

Guinea, Tasmania, Torres

origin of, (111). ■

^^^^^^^^H Morocco, curses in, 361-74.

Straits, Tonga.

Polyandry, adelphic, 301 ; in ,H

Europe, 296. ^^M

^^^^^^H Mortar, 328.

Octave, 24L

^^^^^^H Mother and €hM, 317.

Offerings to rivers, 140,

Poly dae monism, 85. ^^^H

^^^^^^H * Mother of rags/ 122.

Ogam inscriptions. 292.
Old Testament, Folk-lore of.

Polygyny, 58, 350, ^^B

^^^^^^H Motu pottery, 336.

Polynesia, (122), (126), H

^^^^^^^H Mourning customs, no.

101-74.

Polyneflian culture, (126); ^M

^^^^^H Mttkdm,

Omen birds, 179,

family, 311. ■

^^^^^m MMler, Ma3C, (11), (16).
^^^^^^^H Mtiseums, (221) ; of Antliro*

Ophicalcite, 330.

Polyphony, development of, ^|
237. 238. ■

Oracle trom oak, 126.

^^^^^^^H polocrr, 277 ; requisites of,
^^^^H 277 ; types of, 2m-h

Orange, 154.

Polytheism, Greek, 81-100. ^^B

Ordeals, (80), (81), (149).

Ponies of Sigynnae. 261. ^^^M

^^^^^^^H Music, 235-58; function of,

Orenda, 226.

Port Moresby, 326 sq. ^^H

^^^^^^^1 236 ; origin of, 237 ; styles

Orientation, (67).

Pots, breaking, 115. ^^B

^^^^^H

OsciUa, 356.

Pottery,333; club-heads, 335. _^

^^^^^H Musical tone, (37), (114).

Omiah, 227.

Powhatan's mantle, (169). ^^m

^^^^^^^H My cenean god b, 9 L

Oxford, anthropology, (250),

Prayer, 176, ^H

^^^^^^^1 Myekb, C. B„ Ethnological

(257), (258).

Prefaces. (142), (168), (181). ^M

^^^^^H Study of Music, 235-58.

(182), (221), (259). ■

^^^^^^H Mylitta, rite at, 1B9 ; temple

Pachisi, see Lot games.

Prehistoric objects in New ^M

^^^^^H

Paederasty, 303.

GuiQea, 325-43. ■

^^^^^^^1 Mybes, J. L., The Sigynnae,
^^^^^^^H an Ethnological Problem •
^^^^^^B of the Earlylron Age, 255-

Pa^nism, Semitic, 101 sq.
Pairing instinct, 52.
PlEilaeoTithic implements,

Prehistoric races, (21). ^M

Presidential addresses (107^ H
(110), fl25), (146), 2O0), ■
(202), (205). ■

^^^^H

(233).

^^^^^^^H Mystical numhera and music,

Pasturing cattle, 162.

Pnmitive Culture, (30), (38), ■
r41), (42) (53\ (69), (189), ■
(220), (252) ; Society. (43) ; _■

^^^^B

Paternal family, (173), (223).

^^^^H Mythology, (9),(14), (24), (28),

Patolli, see Lot games.

^^^^B (76), (78), (84), (103), (104),
^^^^H (126), (ISB), (137), (149),
^^^^H (180). (192), (211), (231);

Patrilineal descent ofAchae*

thought, 134, ^M

ana, 298.

Procreation, Aastralian be- ^^H

Patronage, 366.

l]efs,21lBq. ^H

^^^^^^^1 Japanese,
^^^^^^^1 Myths, origin of, (9).

Pattems on potteiy, 334.
Pentatonic scale, 242.

Prohibited degrees, 344, ^^M
Propitiation or dead. 100, ^^H

Peredur, 285 sq.

Prostitution, religious, 189sq. ^^H

^^^^^^H Nahal, story of,

Perjury, test of, 135.

Prussian custom, 125. ^^^^

^^^^^H Nameless gods, 83, 91.

^^^^^^^B Name taboos, 91.

Persian custom, 169.
PersoniB cation and animism,

Psychology, (15). ^^^M
Puberty rites, 195. ^^B

^^^^^^^H Nandi ^iistom, 161.

10.

Punaluan family, 310, 315. ^^M

^^^^^^^^H Nasamonian rite, 195 n.

Peru, ofifering to river-god.

Punishment of rape. 363. ^^^|

^^^^^H Natural religion, (180), (188).

140.

Pudeeation. 158. 161, ^^M

^^^^^^H Nature myths, 187.

Pestle, 328, 329 n.

Pygmy beliefs. 227. ^^M

^^^^^^H Neanderthal skull, 66.

P folk, 307.

^^^H

^^^^^^H Negative magic, 219-34.

Phantasms, 3; aa omens, 178.

Quarries, New Guinea, 331, ^^H

^^^^^H Negro dialects, (6).

Philology of f^lang. (57).

Quartertones in music. 245. V

^^^^^^^H Negro-European dialects, (6).

Philosophy of languaj^e, (92);

Queensland beliefs. 2121 ^^^B

^^^^^^H New Guinea archaeology,

of religion, (28); pnmitive,

Questionnaire, 283. ^^H

^^^^^^H 325-43; murder in, 107.

(25).

^^^^1

^^^^^^B New house, 171.

Phonogram, transcription of,

Race to species, relation of, ^^B

^^^^^H

253,

(56). ■

^^^^^^^H Norse custom, 132; in Amer-

Phonograph, manipulation
of. 249-53,

Races of mankind| (116). ^^H

^^^^^H

E&goy 245. ^^H

^^^^^H Nose flute, (126).
^^^^^^H Notes and Queues, antbropo-

Phonology, 307-8.

Rag custom, 369, ^^H

Photography and soul, 145.

Rag-trees, 118. 119, 122; be- ^H

^^^^H logical, (52), (55), (167),

Phratnes, origin of, 55, 206,

hefasto, 129n. ^^B

^^^^^H

Phratry names, 55.

Rainu, 327 sq, ^^H

^^^^^B Numeration, (15), (22), (62),

Picture writing, (137), (149).

Read, C. H„ A Museum of ^^H

^^^^H

Pigment of pottery, 336.
Pilgrimage, 118.

Anthropology. 277-83, ^H
Refugee, 362. 372. ^H

^^^^^^^H OahuE, bregma in, 71.

Filiim, 266; derivation of,

Reindeer, milking, 160. ^1

^^^^^^^1 Oak, 110 sq.; burial under,

266, 269 ; modification of

Relationships, classificatory ^^M

^^^^^^^H 114; in Jewish history,

the, 267.

system, 309-23, ^^B

SUBJECT INDEX

415

Religioii, and magic, 166;
evolution of, 101, (9), (28),
(50). (180). {188), (191);
Syria, 117; of Torres
Straits, 175-88 ; Roman
and Greek compared, 89 ;
nvage, (192).

Religioui proBtitution, 189 sq>

ResearcheB into the Early
History of Mankind and
the Development of Civili-
sation, (5), (7), (26), (94),
(95).

Reverence, 91.

Rnts. Sir J., Nine Witches
of Gloucester, 285-93.

Ehjthm in mimic, 237.

Ridge, frontal, 78.

RiD^EWAY, W-, Who were

the Dorians? 295-308.
lEitual prohibitionB, 154 aq,
[River, crossing, 141.
I^ver-god, 138.
' Rivers, W. H, R., Origin of
the ClasBificatory System
of Relationahipa, 309-23.

Rolle«ton, George, memoir,
(143),
■ Boman gods, 87 ; 9pea.r^ 266.
■Romovet oak of, ISJB.

Roth, W. E., quoted, 213.

Sacred trees, 110 sq.
Sacrifice, 117, 365, 367, 372,

374 ; to ghort, 106.
Sainta, I-^dr for. 368.
Salutation, (160).
Samoa, stone cult, 134,
[Sanction of taboo, 221.
Sajactuariea, Syrian, 118.
Sanskrit, (111
Savage religion, (192);

thought, (24l. (39).
Savages, sexual impulse

among, 53.
Scales, musical, 240.
Scandinavian civilization,

(139).
^ Sc&thach, 288.
Schwalbe on evolution of

human skull, 71 ; on frontal

angle, 66.
Science, primitive, (99).
Scottish skulls, B. N. L an^le

in, 75 J * third point' m,

70.
Scroll pattern, 332.
Sea, sacriBce to, 140.
Seclusion at puberty, 196.
Seething a kid in its mother's

milk, 151.
Seligmann.C. G., Prehistoric

Objects in British New

Guinea, 325-43.
Semitic religion, 101 sq.
Separation of aexeS) 345.

Sequani, 261.

Sex totem, 178,

Sexual character in position
of bregma, 71 ; habits of
horses, 349 ; impulse, 52 ;
intercourse, taboo of, 162 ;
licence, 196; relations of
Spartans, 800; taboo, 232.

Shape shifting, 139.

Sharon, forest of, 111.

Shaving among Spartans,
305 ; protective, 109,

Shechem, oak of, 126.

Shells, carved, S30, E32.

Shrine, totem, 185.

Shrines, hereditary, 178.

Sida, culture-hero, 183.

Sigai, 185.

Sigifnna, 266 sq,, 271,

Sigynnae, characteristics of,
256; of Herodotus, 255-76.

Silures, 289.

Sinew, cutting out, 142.

Skull, Australian, 65 ; divina-
tion, 179,

Slan^, f57J.

Smith, Robertson, 103.

Social condition of man,
primitive, (49) ; conse-
quences of taboo breaking,
228; function of music,
240 ; organization, (43),
(108), (117), (152), (173),
(190), (223); system of
Dorians, 298.

Societies, animal, $46.

Society, primitive, (43).

Somatology of Dorians, 303.

Sonder-Gott, definition of, 87.

*Sonder-Gdtter,' 81-100.

Soul, AustraEan ideas, 213.

Soul-catching, 145.

Soul^keeping, 145,

South Africa, offering to
river-god, 140,

Spain, gaesnm in, 270 ; wea-
pons in, 266.

Sparta, polyandry at, 301 ;
sexual relations in, 300.

Spear-heads, 328 ; in Cyprus,
272 ; of iron age, 268.

Spear-types, 263.

Species, relation of, to race,
(56).

Spells, 176.

Spencer, Herbert, (84), (128).

Spirit, human, 179.

Spirits, in oak-tree, 114; in
threshold, 171.

Spiritualism, 8, (34), (71),
(136), (145),

Spitting custom, 140.

SnitK 245.

Statistical methods, (173),

Status, change of, 320 ; terms
expressive of, 319.

Stillborn children, 172.

Stone age, (54); Tasmania,
(2aS).

Stone implements, 328 ; Tas-
mania (206), <207). (212),
(213), (233), (239).

Stone, in marriage custom,
132 ; oath on, 132.

Stone-throwing, 368, 370.

Stone- worship, 91,

Strabo on the Siginni, 257.

Stranger, intercourse with,
189, 191,200; use of milk
by, 166: taboo of, 232.

Strehlowon the Arunta, 217,

Subordinate divinities, 94.

Sun-god, 187.

Surgery, primitive, 326.

Survivals, 6, 7, (24), (39),
(91), (178); in Jewish reli-
gion, 102.

Sword, Cypriote, 273,

Sympathetic bond, 135, 157 ;
magic, 154, 166; taboo,
220; union, 161.

Taboo, 109, 142,219-34 ; fun-
damentally mystical, 224.

TamasBOS, finds at, 273.

Tamils, ' third point ' in, 70,

Taphian route, 275.

Tarpan, 350.

Tasmania, stone implements,
(206), (207). (212), (213),
(238), (239).

Tasmanians, (181), (182),
(201); as representatives of
palaeolithic man, (207).

Tatu, 106,

Temperament, equal, in
music, 244.

Temple prostitutes, 195,

Terebinth, 120 sq.

Terminology, 4,

Tetrachord, 241.

Thesmophoria, 98.

Thessalians, 303.

■Third point' defined, 69;
position of, 70,

Thomas, N. W., Origin of
Exogamy, 343-54,

Thomson, A., Secret of the
Verge Watch, 355-60.

Threshold, Keepers of the,
167 ; taboos, 167 sq. ; head-
ing on, 173 n.

Thro wing-spear, 265.

Timbre and pitch, 248,

Timor, mourning in, 110,

Titles of gods, 93.

Toda custom, 141, 161,

Todas, widows among, 161.

Tombs at Cnossus, 306.

Tonality, 246.

Tongan kinship, 314.

Torres Straits, kinship of,
312; religion of, 175,

Totem, one, to one totem

416

SUBJECT INDEX

marriage, 205 ; origin of,
55 ; totem posts, (230),
(231), (248); rejztilative of
marriage, 203-10.

Totemism, 9, 54, 103 n., 177,
186. 354. (232); concep-
tional, 217 ; reHct of, in
Murray Islands, 178 ; tran-
sition from, 186,

Tribal brothers, 54 ; marks,
103; relationBhips, 62.

Tribe, origin of, 6i.

Trinity of gode, 128.

Twba, 161.

Two-phratry By* tern, 60»

Tylor, E. B., work of, 1-15,
17, 57, 255, 352, 375-409.

Ukuni, custom of, 155,
Uncircumcised, use of milk

by, 162.
Unit, social, 61,
Unlock, 174,
Unibunna customs, 203,
Usener, Dr., 86 eq,; Odttet^

namen^ 87,

Vampire, 110.

Variability of ekuH, 67, 70,
&c,

Veddahs, * third point ' in, 70,

Vegetable food, 183.

Veneti, 256, 258. 2»8.

Venetia, 258.

Vengeance for death, 103 iq.

Verge watch, 355-9.

Village communitiea, (32).

Vocalization of demonstra-
tives, (8).

Vortigem, 292,

Vows, 117,

Wakamba customs, 162,
Wakikuyu custom. 162.
Wampum belts, (226), (228),
War in Torres Straits, 184.
Warriors, taboos of, 164.
Washing hands, 139.
Watch-cocks, 355-9,
Watchmakers, foreign, 359.
Water-spirit, 138,
Water*throwing, 141.

Weaning, 817.
Well, sacred, 130.
Well-making. 367-
Werwolf, (34).
Westermarck, on exogmmy,

344.
'tVheel, evolution of, (111).
White horses aacrificed, 140,
Widows, purification of, 161;

treatment of, 161.
Wild men, (3).
Winged Eguret of Aaajria,

(184).
Wishing-trees, 115.
Witchcraft, 230.
Witches of Gloucester, 285 sq.
Witches' ladder, (165), (193),
Women and milk, 157; as

dairymaids, 160,
Wounded man and milk,

158.
Tao belief, 105 ; custom,

105,

Zoffo, 176, 185.

IL INDEX OF GREEK WORDS

axpaT(m6Trjtf 82, 94.
*AXx/dai. 85.
dfi(pidpQfioi, 82, 95.
apa^pidttf 259.
Al^Tjaiat 81.

fiXavTrif 82, 95,

TtPtrvWldtSf 83,

d^doLj^or, 91,

Aal^ovfri 81, 82, 85 ; ttrri-

Aaif^e^Vf meaning of, 85 ; ^m*
dtidnjiT, 81, 94 ; i(TodaiTTjSt 82.
Aai'njf, 82, 94.
Aavdiatf 308,
AttjryivSf 94.

*EwifiaXi^f, 83.
*EirtT*y40r, 82, 95.
*Ep*x^(it* 96.
'Ept>ta, 81,98.
'Erij^tXo, 83,

EinddMfCOi, 83.
Eirdpojuor, 82, 96,
EvvoaToSj 81, 95.
*ExfrXaIof, 81, 95,

"Hpwrj 81 sq. ; farpot-, 82, 97 ,
tcara irpvfi*>air, 82, 97 ; tttftn-
fxQff 82 ; TTpit itv\q>v, 82 ;
tn'eiPnifri(fy6pOit 82 ; urparrj'
y6t, 82, 97,

OaXX<t>, 81.

Bfol 'AyopoiOi, 84 ; ^Anorpn-
rratoi, 83 ; ica$apmj 83 ;
MnXixtcM, 84 ; Mii'Xafrftut,
84 ; Upa^i^iKaif 84 ; IIpo-
dpoptU^ 84 ; ^prfTptoif 84,

ftfodaiTTjr, 82, 98,
mwoSj 307.

KoXXtyt I'fici, 82, 98,

Kapnoif 81.
Ktpdtiiv, 82, 94,

K\aUott>^pot, 82, 97.
Kfwporp6<Potj 82, 98.
KvafUTTjs^ 81, 96.
K6>Xtad«c, 83.

Maidot, 260.
MarrwF, 82, 94.
Mf-yMTrof, 92.
M«Xi;^tor, 94.
Mvurypor, 83, 98.

^yiK^f Aai^w, 98.

Xrixidni, 308.
trt^viinov, 269.
trtyvmfOff 261.
ffWQV&atuv ^aifuap^ 82, 95.

Tapditwnou 83, 97.
Tfixc^i/Xaf , 82.
rcXfcridpo/inr. 82*

TfTTapfff, 307.
tpapfAaKi^iS^ 84.

Oxford : Printed at the Clarendon Press by Horaok Hart^ M. A,

■

^^^^^^^^^^H
^^^^^^^^^^^^^B.

Bl4 u ^^

^^^^1 )

^' nor*' - *

«,,^ APR 05 1993 ■

Stanford University library J

Stanford, Califonita ^H

In order that others may use th w book.
please return it as soon as possible, but _
not later than the date due, 1

Internet Archive Audio

Featured

Top

Images

Featured

Top

Software

Featured

Top

Texts

Featured

Top

Video

Featured

Top

Mobile Apps

Browser Extensions

Archive-It Subscription

Save Page Now

Full text of "Anthropological essays presented to Edward Burnett Tylor in honour of his 75th birthday, October 2, 1907"

See other formats